Elevation of blood urea nitrogen is predictive of long-term mortality in critically ill patients independent of "normal" creatinine

Construction and validation of a predictive model of mortality of tuberculosis-destroyed lung patients requiring mechanical ventilation: A single-center retrospective cohort study

The mortality rate for intensive care unit tuberculosis-destroyed lung (TDL) patients requiring mechanical ventilation (MV) remains high. We conducted a retrospective analysis of adult TDL patients requiring MV who were admitted to the intensive care unit of a tertiary infectious disease hospital in Chengdu, Sichuan Province, China from January 2019 to March 2023. Univariate and multivariate COX regression analyses were conducted to determine independent patient prognostic risk factors that were used to construct a predictive model of patient mortality. A total of 331 patients were included, the median age was 63.0 (50.0-71.0) years, 262 (79.2%) were males and the mortality rate was 48.64% (161/331). Training and validation data sets were obtained from 245 and 86 patients, respectively. Analysis of the training data set revealed that body mass index <18.5 kg/m2, blood urea nitrogen ≥7.14 mmol/L and septic shock were independent risk factors for increased mortality of TDL patients requiring MV. These variables were then used to construct a risk-based model for predicting patient mortality. Area under curve, sensitivity, and specificity values obtained using the model for the training data set were 0.808, 79.17%, and 68.80%, respectively, and corresponding values obtained using the validation data set were 0.876, 95.12%, and 62.22%, respectively. Concurrent correction curve and decision curve analyses confirmed the high predictive ability of the model, indicating its potential to facilitate early identification and classification-based clinical management of high-risk TDL patients requiring MV.

Association between serum osmolality and risk of in-hospital mortality in patients with intracerebral hemorrhage

Aim

This study aimed to analyze the association between serum osmolality and the risk of in-hospital mortality in intracerebral hemorrhage (ICH) patients.

Methods

In this retrospective cohort study, data of a total of 1,837 ICH patients aged ≥18 years were extracted from the Medical Information Mart for Intensive Care-IV (MIMIC-IV). Serum osmolality and blood urea nitrogen (BUN)-to-creatinine (Cr) ratio (BCR) were used as the main variables to assess their association with the risk of in-hospital mortality in ICH patients after first intensive care unit (ICU) admission using a univariable Cox model. Univariable and multivariable Cox regression analyses were applied to explore the associations between serum osmolality, BCR, and in-hospital mortality of ICH patients. Hazard ratio (HR) and 95% confidence intervals (CIs) were calculated.

Results

The median survival duration of all participants was 8.29 (4.61-15.24) days. Serum osmolality of ≥295 mmol/L was correlated with an increased risk of in-hospital mortality in patients with ICH (HR = 1.43, 95%CI: 1.14-1.78). BCR of >20 was not significantly associated with the risk of in-hospital mortality in ICH patients. A subgroup analysis indicated an increased risk of in-hospital mortality among ICH patients who were women, belonged to white or Black race, or had complications with acute kidney injury (AKI).

Conclusion

High serum osmolality was associated with an increased risk of in-hospital mortality among ICH patients.

Serum urea increase during hospital stay is associated with worse outcomes after in-hospital cardiac arrest

BACKGROUND

Evaluate the association between serum urea at admission and during hospital stay with return of spontaneous circulation (ROSC) and in-hospital mortality in patients with in-hospital cardiac arrest (IHCA).

METHODS

This retrospective study included patients over 18 years with IHCA attended from May 2018 to December 2022. The exclusion criteria were the absence of exams to calculate delta urea and the express order of "do-not-resuscitate". Data were collected from the electronic medical records. Serum admission urea and urea 24 hours before IHCA were also collected and used to calculate delta urea.

RESULTS

A total of 504 patients were evaluated; 125 patients were excluded due to the absence of variables to calculate delta urea and 5 due to "do-not-resuscitate" order. Thus, we included 374 patients in the analysis. The mean age was 65.0 ± 14.5 years, 48.9% were male, 45.5% had ROSC, and in-hospital mortality was 91.7%. In logistic regression models, ROSC was associated with lower urea levels 24 hours before IHCA (OR: 0.996; CI95%: 0.992-1.000; p: 0.032). In addition, increased levels of urea 24 hours before IHCA (OR: 1.020; CI95%: 1.008-1.033; p: 0.002) and of delta urea (OR: 1.001; CI95%: 1.001-1.019; p: 0.023) were associated with in-hospital mortality. ROC curve analysis showed that the area under the ROC curve for mortality prediction was higher for urea 24 hours before IHCA (Cutoff > 120.1 mg/dL) than for delta urea (Cutoff > 34.83 mg/dL).

CONCLUSIONS

In conclusion, increased serum urea levels during hospital stay were associated with worse prognosis in IHCA.

Mortality of elderly patients with acute kidney injury undergoing continuous renal replacement therapy: is age a risk factor?

BACKGROUND

Whether advanced age is associated with poor outcomes of elderly patients with acute kidney injury (AKI) requiring continuous renal replacement therapy (CRRT) is controversial. This study aimed to evaluate age effect and predictors for mortality in elderly AKI patients undergoing CRRT.

METHODS

Data of 480 elderly AKI patients who underwent CRRT were retrospectively analyzed. Subjects were stratified into two groups according to age: younger-old (age, 65-74 years; n = 205) and older-old (age, ≥75 years; n = 275). Predictors for 28-day and 90-day mortality and age effects were analyzed using multivariable Cox regression analysis and propensity score matching.

RESULTS

Urine output at the start of CRRT (adjusted hazard ratio [aHR], 0.99; 95% confidence interval [CI], 0.99-1.00; p = 0.04), operation (aHR, 0.53; 95% CI, 0.30-0.93; p = 0.03), and use of an intra-aortic balloon pump (aHR, 3.60; 95% CI, 1.18-10.96; p = 0.02) were predictors for 28-day mortality. Ischemic heart disease (aHR, 1.74; 95% CI, 1.02-2.98; p = 0.04) and use of a ventilator (aHR, 0.56; 95% CI, 0.36-0.89; p = 0.01) were predictors for 90-day mortality. The older-old group did not exhibit a higher risk for 28- day or 90-day mortality than the younger-old group in multivariable or propensity score-matched models.

CONCLUSION

Advanced age was not a risk factor for mortality among elderly AKI patients undergoing CRRT, suggesting that advanced age should not be considered for therapeutic decisions in critically ill elderly patients with AKI requiring CRRT.

A novel prognostic model to predict mortality in patients with acute-on-chronic liver failure in intensive care unit

Acute-on-chronic liver failure (ACLF) implies high short-term mortality rates and usually requires intensive care unit (ICU) admission. Proper prognosis for these patients is crucial for early referral for liver transplantation. The superiority of CLIF-C ACLF score in Asian patients with ACLF admitted to an ICU remains inconclusive when compared to other scoring systems. The purpose of the study is (i) to compare the predictive performance of original MELD, MELD-Lactate, CLIF-C ACLF, CLIF-C ACLF-Lactate, and APACHE-II scores for short-term mortality assessment. (ii) to build and validate a novel scoring system and to compare its predictive performance to that of the original five scores. Two hundred sixty-five consecutive cirrhotic patients with ACLF who were admitted to our ICU were enrolled. The prognostic values for mortality were assessed by ROC analysis. A novel model was developed and internally validated using fivefold cross-validation. Alcohol abuse was identified as the primary etiology of cirrhosis. The AUROC of the five prognostic scores were not significantly superior to each other in predicting 1-month and 3-month mortality. The newly developed prognostic model, incorporating age, alveolar-arterial gradient (A-a gradient), BUN, total bilirubin level, INR, and HE grades, exhibited significantly improved performance in predicting 1-month and 3-month mortality with AUROC of 0.863 and 0.829, respectively, as compared to the original five prognostic scores. The novel ACLF model seems to be superior to the original five scores in predicting short-term mortality in ACLF patients admitted to an ICU. Further rigorous validation is required.

Optimizing sepsis treatment strategies via a reinforcement learning model

Purpose

The existing sepsis treatment lacks effective reference and relies too much on the experience of clinicians. Therefore, we used the reinforcement learning model to build an assisted model for the sepsis medication treatment.

Methods

Using the latest Sepsis 3.0 diagnostic criteria, 19,582 sepsis patients were screened from the Medical Intensive Care Information III database for treatment strategy research, and forty-six features were used in modeling. The study object of the medication strategy is the dosage of vasopressor drugs and intravenous infusion. Dueling DDQN is proposed to predict the patient's medication strategy (vasopressor and intravenous infusion dosage) through the relationship between the patient's state, reward function, and medication action. We also constructed protection against the possible high-risk behaviors of Dueling DDQN, especially sudden dose changes of vasopressors can lead to harmful clinical effects. In order to improve the guiding effect of clinically effective medication strategies on the model, we proposed a hybrid model (safe-dueling DDQN + expert strategies) to optimize medication strategies.

Results

The Dueling DDQN medication model for sepsis patients is superior to clinical strategies and other models in terms of off-policy evaluation values and mortality, and reduced the mortality of clinical strategies from 16.8 to 13.8%. Safe-Dueling DDQN we proposed, compared with Dueling DDQN, has an overall reduction in actions involving vasopressors and reduces large dose fluctuations. The hybrid model we proposed can switch between expert strategies and safe dueling DDQN strategies based on the current state of patients.

Conclusions

The reinforcement learning model we proposed for sepsis medication treatment, has practical clinical value and can improve the survival rate of patients to a certain extent while ensuring the balance and safety of medication.

A High Preoperative Blood Urea Nitrogen to Serum Albumin Ratio Does Not Predict Worse Outcomes Following the Robotic-Assisted Pulmonary Lobectomy for Lung Cancer

BACKGROUND

 The blood urea nitrogen to serum albumin ratio (BAR) is an emerging prognostic parameter of interest. The utility of BAR as a prognostic factor has not been analyzed in lung cancer patients undergoing pulmonary lobectomy. We evaluated the ability of High BAR to predict worse outcomes after robotic-assisted pulmonary lobectomy (RAPL) for lung cancer.

METHODS

 We retrospectively analyzed 400 patients who underwent RAPL from September 2010 to March 2022 by one surgeon. Patients were stratified by Low BAR (<6.25 mg/g) and High BAR (≥6.25 mg/g). Patients' demographics, tumor characteristics, comorbidities, surgical complications, outcomes, and survival were collected and compared by High and Low BAR groups. The primary outcome of interest was 30-day mortality.

RESULTS

 Receiver operator curves (ROC) confirmed that 6.25 was an optimal threshold for estimating mortality based on Low and High BAR. There were no differences in surgical complications or outcomes between the Low and High BAR groups. The ability of BAR to predict 30-day mortality was evaluated with the area under the curve (AUC) analysis, which showed that higher BAR could not predict mortality (AUC=0.655; 95% CI, 0.435-0.875; p=0.166). Similarly, survival analysis revealed no difference in five-year overall survival between the Low and High BAR groups (p=0.079).

CONCLUSION

 High BAR did not predict worse outcomes after RAPL for lung cancer in our study. Further studies are needed to better determine the prognostic ability of BAR in lower-risk populations.

Clinical outcome prediction using observational supervision with electronic health records and audit logs

OBJECTIVE

Audit logs in electronic health record (EHR) systems capture interactions of providers with clinical data. We determine if machine learning (ML) models trained using audit logs in conjunction with clinical data ("observational supervision") outperform ML models trained using clinical data alone in clinical outcome prediction tasks, and whether they are more robust to temporal distribution shifts in the data.

MATERIALS AND METHODS

Using clinical and audit log data from Stanford Healthcare, we trained and evaluated various ML models including logistic regression, support vector machine (SVM) classifiers, neural networks, random forests, and gradient boosted machines (GBMs) on clinical EHR data, with and without audit logs for two clinical outcome prediction tasks: major adverse kidney events within 120 days of ICU admission (MAKE-120) in acute kidney injury (AKI) patients and 30-day readmission in acute stroke patients. We further tested the best performing models using patient data acquired during different time-intervals to evaluate the impact of temporal distribution shifts on model performance.

RESULTS

Performance generally improved for all models when trained with clinical EHR data and audit log data compared with those trained with only clinical EHR data, with GBMs tending to have the overall best performance. GBMs trained with clinical EHR data and audit logs outperformed GBMs trained without audit logs in both clinical outcome prediction tasks: AUROC 0.88 (95% CI: 0.85-0.91) vs. 0.79 (95% CI: 0.77-0.81), respectively, for MAKE-120 prediction in AKI patients, and AUROC 0.74 (95% CI: 0.71-0.77) vs. 0.63 (95% CI: 0.62-0.64), respectively, for 30-day readmission prediction in acute stroke patients. The performance of GBM models trained using audit log and clinical data degraded less in later time-intervals than models trained using only clinical data.

CONCLUSION

Observational supervision with audit logs improved the performance of ML models trained to predict important clinical outcomes in patients with AKI and acute stroke, and improved robustness to temporal distribution shifts.

Association between the first 24 hours PaCO2 and all-cause mortality of patients suffering from sepsis-associated encephalopathy after ICU admission: A retrospective study

OBJECTIVE

The relationship between the levels of the first 24-h PaCO2 and the prognosis of sepsis-associated encephalopathy (SAE) remains unclear, and the first 24-h optimal target for PaCO2 is currently inconclusive. This study was performed to investigate the correlation between PaCO2 and all-cause mortality for SAE patients, establish a reference range of the initial 24-hour PaCO2 for clinicians in critical care, and explain the possible pathophysiological mechanisms of abnormal PaCO2 levels as a higher mortality risk factor for SAE.

METHODS

The baseline information and clinical data of patients were extracted from the fourth edition Medical Information Mart for Intensive Care database (MIMIC-IV 2.0). Multivariate logistic regressions were performed to assess the relationship between PaCO2 and all-cause mortality of SAE. Additionally, restricted cubic splines, Kaplan-Meier Survival analyses, propensity score matching (PSM) analyses, and subgroup analyses were conducted.

RESULTS

A total of 5471 patients were included in our cohort. In the original and matched cohort, multivariate logistic regression analysis showed that normocapnia and mild hypercapnia may be associated with a more favorable prognosis of SAE patients, and survival analysis supported the findings. In addition, a U-shaped association emerged when examining the initial 24-hour PaCO2 levels in relation to 30-day, 60-day, and 90-day mortality using restricted cubic splines, with an average cut-off value of 36.3mmHg (P for nonlinearity<0.05). Below the cut-off value, higher PaCO2 was associated with lower all-cause mortality, while above the cut-off value, higher PaCO2 was associated with higher all-cause mortality. Subsequent subgroup analyses revealed similar results for the subcohort of GCS≤8 compared to the original cohort. Additionally, when examining the subcohort of GCS>8, a L-shaped relationship between PaCO2 and the three clinical endpoints emerged, in contrast to the previously observed U-shaped pattern. The findings from the subcohort of GCS>8 suggested that patients experiencing hypocapnia had a more unfavorable prognosis, which aligns with the results obtained from corresponding multivariate logistic regression analyses.

CONCLUSION

The retrospective study revealed the association between the first 24-h PaCO2 and all-cause mortality risk (30-day, 60-day, and 90-day) for patients with SAE in ICU. The range (35mmHg-50mmHg) of PaCO2 may be the optimal target for patients with SAE in clinical practice.

Prognostic Usefulness of Basic Analytical Data in Chronic Obstructive Pulmonary Disease Exacerbation

Introduction

COPD causes high morbidity and mortality and high health costs. Thus, identifying and analyzing the distinctive and treatable traits seems useful to optimize the management of AEPOC patients. While various biomarkers have been researched, no solid data for systematic use have been made available.

Aim

Assessing the short-term prognostic usefulness of clinical and analytical parameters available in routine clinical practice in COPD exacerbations.

Material and methods

Multicenter prospective observational study conducted between 2016 and 2018. Patients admitted for COPD exacerbation who agreed to participate and signed an informed consent form were included. Prolonged stay, in-hospital mortality or early readmission was considered an unfavorable progression. 30-Day mortality was also analyzed.

Results

615 patients were included. Mean age was 73.9 years (SD 10.6); 86.2% were male. Progression of 357 patients (58%) was considered unfavorable. Mortality at 1 month from discharge was 6.7%. The multivariate analysis shows a relationship between the CRP/Albumin ratio and unfavorable progression (OR 1.008, 95% CI 1.00; 1.01), as well as increased risk of death at 1 month from discharge with elevated urea (OR 1.01, 95% CI 1.005; 1.02) and troponin T (OR 2.21, 95% CI 1.06; 4.62).

Conclusion

Elevated CRP/Albumin, urea and TnT are prognostic indicators of poor short-term outcome in patients admitted for COPD exacerbation. Cardiovascular comorbidity and systemic inflammation could explain these findings.

Clinical Manifestations and Risk Factors of Tigecycline-Associated Thrombocytopenia

Background

Thrombocytopenia, characterized by a diminished platelet count, emerged as the most frequently reported coagulation dysfunction event according to the FDA Adverse Event Reporting System (FAERS) database. In recent years, numerous clinical studies have investigated the potential link between tigecycline usage and the occurrence of hypofibrinogenemia. However, a research gap remains in comprehensively examining the association between tigecycline and thrombocytopenia in real-world settings.

Methods

This study was conducted to explore the incidence and clinical manifestations of tigecycline-associated thrombocytopenia. A retrospective case-control study of patients treated with tigecycline was conducted between January 2018 and June 2022.

Results

In total, 373 patients were included in this study. Among these patients, 12.3% experienced thrombocytopenia. The onset of thrombocytopenia occurred within a range of 2 to 22 days after the initiation of tigecycline, with a median period (25-75th percentile) of 9 (6-11) days. Among the patients manifesting thrombocytopenia, 60.9% exhibited mild-to-moderate cases (grades 1-2) while 39.1% endured severe cases (grades 3-4). Multivariate analysis delineated several factors as independent risk factors for thrombocytopenia. Notably, advanced age (≥74 years) (p=0.028), risk of malnutrition (p<0.001), tigecycline therapy for ≥7 days (p=0.003), DBIL>8.1μmol/L (p<0.001)), BUN>8.1mmol/L (p=0.002) emerged as independent risk factors associated with thrombocytopenia. When comparing the control group to the thrombocytopenia group, 70.7% of patients in the control group exhibited 0-2 risk factors, while all patients in the thrombocytopenia group demonstrated risk factors. Specifically, 95.7% of patients in the thrombocytopenia group presented with three to five risk factors, with only 4.4% having 0-2 risk factors.

Conclusion

Tigecycline administration is associated with thrombocytopenia. Healthcare professionals should exercise vigilance, particularly in cases of severe tigecycline-associated thrombocytopenia, and undertake routine monitoring of patients' platelet counts, especially for those who possess three or more risk factors.

Prognostic value of blood urea nitrogen to serum albumin ratio for acute kidney injury and in-hospital mortality in intensive care unit patients with intracerebral haemorrhage: a retrospective cohort study using the MIMIC-IV database

OBJECTIVE

We sought to evaluate the prognostic ability of blood urea nitrogen to serum albumin ratio (BAR) for acute kidney injury (AKI) and in-hospital mortality in patients with intracerebral haemorrhage (ICH) in intensive care unit (ICU).

DESIGN

A retrospective cohort study using propensity score matching.

SETTING

ICU of Beth Israel Deaconess Medical Center.

PARTICIPANTS

The data of patients with ICH were obtained from the Medical Information Mart for Intensive Care IV (V.1.0) database. A total of 1510 patients with ICH were enrolled in our study.

MAIN OUTCOME AND MEASURE

The optimal threshold value of BAR is determined by the means of X-tile software (V.3.6.1) and the crude cohort was categorised into two groups on the foundation of the optimal cut-off BAR (6.0 mg/g). Propensity score matching and inverse probability of treatment weighting were performed to control for confounders. The predictive performance of BAR for AKI was tested using univariate and multivariate logistic regression analyses. Multivariate Cox regression analysis was used to investigate the association between BAR and in-hospital mortality.

RESULTS

The optimal cut-off value for BAR was 6.0 mg/g. After matching, multivariate logistic analysis showed that the high-BAR group had a significantly higher risk of AKI (OR, 2.60; 95% confidence index, 95% CI, 1.86 to 3.65, p<0.001). What's more, a higher BAR was also an independent risk factor for in-hospital mortality (HR, 2.84; 95% confidence index, 95% CI, 1.96 to 4.14, p<0.001) in terms of multivariate Cox regression analysis. These findings were further demonstrated in the validation cohort.

CONCLUSIONS

BAR is a promising and easily available biomarker that could serve as a prognostic predictor of AKI and in-hospital mortality in patients with ICH in the ICU.

Higher urea-to-albumin ratio is associated with mortality risk in critically ill COVID-19 patients

BACKGROUND

This study aimed to evaluate the ability of the urea-to-albumin ratio (UAR) to predict mortality in critically ill coronavirus disease 2019 (COVID-19) patients.

METHODS

This retrospective study included adult patients admitted with COVID-19 at two intensive care units (ICUs) at the University Hospital. Serum urea and albumin concentrations at ICU admission were used to calculate the UAR. All patients were followed up during hospitalization, and the ICU mortality rate was recorded.

RESULTS

Two hundred and eleven patients were evaluated. The mean age was 57.8 ± 15.5 years, and 54% were male. Approximately 84.4% of patients were considered to be at nutritional risk by the NRS 2002, and the median UAR was 18.3 (10.5-34.8). The length of stay in the ICU was 10 (6-16) days, 38.4% of the patients required dialysis, and 64.9% died. Age, male sex, need of hemodialysis, lactate level, and inflammatory parameters were associated with higher mortality. Patients non-survivors had a higher UAR (23.7 [13.6-41.8] vs. 10.9 [8.5-16.8]; p < 0.001). The cutoff point with the best performance of UAR in the ROC curve for predicting mortality was ≥12.17 (AUC: 0.7201; CI 95%: 0.656-0.784). Additionally, the risk of mortality was 2.00-fold in the group of patients with UAR ≥12.17 (HR: 2.00 CI: 1.274-3.149; p = 0.003) and remained significant after adjusted analyzes (models 1 and 2).

CONCLUSION

Our data suggest that a UAR ≥12.17 increased the risk of mortality by 2.00-fold in critically ill COVID-19 patients.

A PREDICTION MODEL FOR SEPSIS IN INFECTED PATIENTS: EARLY ASSESSMENT OF SEPSIS ENGAGEMENT

ABSTRACT

Purpose: To evaluate significant risk variables for sepsis incidence and develop a predictive model for rapid screening and diagnosis of sepsis in patients from the emergency department (ED). Methods: Sepsis-related risk variables were screened based on the PIRO (Predisposition, Insult, Response, Organ dysfunction) system. Training (n = 1,272) and external validation (n = 568) datasets were collected from Peking Union Medical College Hospital (PUMCH) and Beijing Tsinghua Changgung Hospital (BTCH), respectively. Variables were collected at the time of admission. Sepsis incidences were determined within 72 h after ED admissions. A predictive model, Early Assessment of Sepsis Engagement (EASE), was developed, and an EASE-based nomogram was generated for clinical applications. The predictive ability of EASE was evaluated and compared with the National Early Warning Score (NEWS) scoring system. In addition, internal and external validations were performed. Results: A total of 48 characteristics were identified. The EASE model, which consists of alcohol consumption, lung infection, temperature, respiration rate, heart rate, serum urea nitrogen, and white blood cell count, had an excellent predictive performance. The EASE-based nomogram showed a significantly higher area under curve (AUC) value of 86.5% (95% CI, 84.2%-88.8%) compared with the AUC value of 78.2% for the NEWS scoring system. The AUC of EASE in the external validation dataset was 72.2% (95% CI, 66.6%-77.7%). Both calibration curves of EASE in training and external validation datasets were close to the ideal model and were well-calibrated. Conclusions: The EASE model can predict and screen ED-admitted patients with sepsis. It demonstrated superior diagnostic performance and clinical application promise by external validation and in-parallel comparison with the NEWS scoring system.

BUN level is associated with cancer prevalence

Blood urea nitrogen (BUN) was an important biomarker for the development and prognosis of many diseases. Numerous studies had demonstrated that BUN had a strong relationship with long-term mortality, survival and the prevalence of some diseases. The diagnosis and treatment, prognosis and long-term survival rate of cancer were the focus of clinical research at present. However, the relationship between BUN level and cancer prevalence was not clear. To investigate the relationship between BUN level and cancer prevalence, we performed a statistical analysis of population data from the National Health and Nutrition Examination Survey (NHANES) database. The results of the study showed that BUN level were positively correlated with cancer prevalence, and the correlation was more pronounced in breast cancer.

Serum Urea-to-Albumin Ratio Is an Independent Predictor of Intra-Hospital Mortality in Neurosurgical Intensive Care Unit Patients with Spontaneous Intracerebral Hemorrhage

The negative prognostic value of an increased serum urea-to-albumin ratio on intra-hospital mortality is frequently investigated in general critically ill patients and patients with septic shock, although not in neurosurgical patients with spontaneous intracerebral hemorrhages (ICH). The current study was conducted to investigate the impact of the serum urea-to-albumin ratio upon hospital admission on intra-hospital mortality in ICU-admitted neurosurgical patients with spontaneous ICH.

METHODS

This retrospective study analyzed 354 ICH patients, who were treated from 10/2008 to 12/2017 at our intensive care units (ICU). Blood samples were taken upon admission, and the patients' demographic, medical, and radiological data were analyzed. A binary logistic regression analysis was performed for the identification of independent prognostic parameters for intra-hospital mortality.

RESULTS

Overall, the intra-hospital mortality rate was 31.4% (n = 111). In the binary logistic analysis, a higher serum urea-to-albumin ratio (OR = 1.9, CI = 1.23-3.04, p = 0.005) upon admission was identified as an independent predictor of intra-hospital mortality. Furthermore, a serum urea-to-albumin ratio cut-off level of >0.01 was associated with raised intra-hospital mortality (Youden's index = 0.32, sensitivity = 0.57, specificity = 0.25).

CONCLUSION

A serum urea-to-albumin ratio greater than 1.1 seems to be a prognostic marker to predict intra-hospital mortality in patients with ICH.

Risk Factors for Death Among Veterans Following Acute Kidney Injury

BACKGROUND

Acute kidney injury is prevalent among hospitalized veterans, and associated with increased risk of death following discharge. However, risk factors for death following acute kidney injury have not been well defined. We developed a mortality prediction model using Veterans Health Administration data.

METHODS

This retrospective cohort study included inpatients from 2013 through 2018 with a creatinine increase of ≥0.3 mg/dL. We evaluated 45 variables for inclusion in our final model, with a primary outcome of 1-year mortality. Bootstrap sampling with replacement was used to identify variables selected in >60% of models using stepwise selection. Best sub-sets regression using Akaike information criteria was used to identify the best-fitting parsimonious model.

RESULTS

A total of 182,683 patients were included, and 38,940 (21.3%) died within 1 year of discharge. The 10-variable model to predict mortality included age, chronic lung disease, cancer within 5 years, unexplained weight loss, dementia, congestive heart failure, hematocrit, blood urea nitrogen, bilirubin, and albumin. Notably, acute kidney injury stage, chronic kidney disease, discharge creatinine, and proteinuria were not selected for inclusion. C-statistics in the primary validation cohorts were 0.77 for the final parsimonious model, compared with 0.52 for acute kidney injury stage alone.

CONCLUSION

We identified risk factors for long-term mortality following acute kidney injury. Our 10-variable model did not include traditional renal variables, suggesting that non-kidney factors contribute to the risk of death more than measures of kidney disease in this population, a finding that may have implications for post-acute kidney injury care.

Development and validation of a nomogram for predicting in-hospital mortality of elderly patients with persistent sepsis-associated acute kidney injury in intensive care units: a retrospective cohort study using the MIMIC-IV database

OBJECTIVES

To identify the clinical risk factors that influence in-hospital mortality in elderly patients with persistent sepsis-associated acute kidney injury (S-AKI) and to establish and validate a nomogram to predict in-hospital mortality.

DESIGN

Retrospective cohort analysis.

SETTING

Data from critically ill patients at a US centre between 2008 and 2021 were extracted from the Medical Information Mart for Intensive Care (MIMIC)-IV database (V.1.0).

PARTICIPANTS

Data from 1519 patients with persistent S-AKI were extracted from the MIMIC-IV database.

PRIMARY OUTCOME

All-cause in-hospital death from persistent S-AKI.

RESULTS

Multiple logistic regression revealed that gender (OR 0.63, 95% CI 0.45-0.88), cancer (2.5, 1.69-3.71), respiratory rate (1.06, 1.01-1.12), AKI stage (2.01, 1.24-3.24), blood urea nitrogen (1.01, 1.01-1.02), Glasgow Coma Scale score (0.75, 0.70-0.81), mechanical ventilation (1.57, 1.01-2.46) and continuous renal replacement therapy within 48 hours (9.97, 3.39-33.9) were independent risk factors for mortality from persistent S-AKI. The consistency indices of the prediction and the validation cohorts were 0.780 (95% CI: 0.75-0.82) and 0.80 (95% CI: 0.75-0.85), respectively. The model's calibration plot suggested excellent consistency between the predicted and actual probabilities.

CONCLUSIONS

This study's prediction model demonstrated good discrimination and calibration abilities to predict in-hospital mortality of elderly patients with persistent S-AKI, although it warrants further external validation to verify its accuracy and applicability.

Blood urea nitrogen - independent marker of mortality in sepsis

BACKGROUND

This retrospective study examines the relationship between admission Blood Urea Nitrogen (BUN) levels and clinical outcomes in patients with sepsis from two separate cohorts in the Czech Republic and the United States.

METHODS

The study included 9126 patients with sepsis between January 2014 and December 2018. Kaplan-Meier survival curves and Cox regression were used to analyse the data. An optimal cut-off was calculated by means of the Youden-Index.

RESULTS

BUN at ICU admission was categorized as 10-20, 20-40 and >40 mg/dL. Comparing the group with the highest BUN levels to the one with lowest levels, we found HR for 28 days mortality 2.764 (CI 95% 2.37-3.20; P<0.001). We derived an optimal cut-off for prediction of 28 days mortality of 23 mg/dL. The association between BUN and 28 days mortality remained significant after adjusting for potential confounders - for APACHE IV (HR 1.374; 95%CI 1.20-1.58; P<0.001), SAPS2 (HR 1.545; 95%CI 1.35-1.77; P<0.001), eGFR (HR 1.851; 95%CI 1.59-2.16; P<0.001) and several other variables in an integrative model.

CONCLUSIONS

Our findings support the BUN level as an independent and easily available predictor of 28 days mortality in septic critically ill patients admitted to an ICU.

The National Early Warning Score 2(NEWS2) to Predict Early Progression to Severe Community-Acquired Pneumonia

This study aimed to assess the predictive performance of the National Early Warning Score 2 (NEWS2) to identify the early progression to severe disease in patients with community-acquired pneumonia (CAP). A prospective-cohort study was conducted among patients with CAP admitted to a university hospital between October 2020 and December 2021. The endpoint of interest was the progression to severe CAP, defined as the requirement for a mechanical ventilator, a vasopressor, or death within 72 h after hospital admission. Among 260 patients, 53 (25.6%) had early progression to severe CAP. The median NEWS2 of the early progression group was higher than that of the non-progression group [8 (6-9) vs. 7 (5-8), p = 0.015, respectively]. The AUROC of NEWS2 to predict early progression to severe CAP was 0.61 (95% CI: 0.52-0.70), while IDSA/ATS minor criteria ≥ 3 had AUROC 0.56 (95% CI 0.48-0.65). The combination of NEWS2 ≥ 8, albumin level < 3 g/dL and BUN ≥ 30 mg/dL improved AUROC from 0.61 to 0.71 (p = 0.015). NEWS2 and IDSA/ATS minor criteria showed fair predictive-accuracy in predicting progression to severe CAP. The NEWS2 cut-off ≥ 8 in combination with low albumin and uremia improved predictive-accuracy, and could be easily used in general practice.

Effect of withdrawal of thyroid hormones versus administration of recombinant human thyroid-stimulating hormone on renal function in thyroid cancer patients

This study was conducted to investigate the effects of thyroid hormone withdrawal (THW) and recombinant human thyroid-stimulating hormone (rhTSH) administration on renal function in patients with thyroid cancer after total thyroidectomy. This study included 202 patients who discontinued thyroid hormone therapy and/or received rhTSH after total thyroidectomy. Creatinine (Cr), blood urea nitrogen (BUN) levels, and estimated glomerular filtration rate (eGFR) were assessed at the following three time points: before thyroidectomy, at least 3 weeks after THW, and 1 day after the second injection of rhTSH. The median serum Cr level was significantly higher following THW compared to that before thyroidectomy (0.95 versus 0.70). In contrast, the median BUN level was significantly lower after THW compared to that before thyroidectomy (9.8 versus 11.3). Over a fifth (22.2%) of patients had abnormal eGFR values after THW, which was significantly greater than that before thyroidectomy. In contrast, renal parameter values after rhTSH administration were not significantly different than those before thyroidectomy. In conclusion, THW affects renal function in patients with thyroid cancer who have undergone total thyroidectomy. However, renal function in such patients is not affected by rhTSH administration.

Blood urea nitrogen to serum albumin ratio: a good predictor of in-hospital and 90-day all-cause mortality in patients with acute exacerbations of chronic obstructive pulmonary disease

BACKGROUND

Previous studies on acute exacerbation of chronic obstructive pulmonary disease (AECOPD) have found that those who died in hospital had higher blood urea nitrogen levels and a worse nutritional status compared to survivors. However, the association between the blood urea nitrogen to serum albumin ratio (BUN/ALB ratio) and in-hospital and short-term prognosis in patients with AECOPD remains unclear. The aim of this study was to explore the usefulness of BUN/ALB ratio in AECOPD as an objective predictor for in-hospital and 90-day all-cause mortality.

METHODS

We recorded the laboratory and clinical data in patients with AECOPD on admission. By drawing the ROC curve for the patients, we obtained the cut-off point for the BUN/ALB ratio for in-hospital death. Multivariate logistic regression was used for analyses of the factors of in-hospital mortality and multivariate Cox regression was used to analyze the factors of 90-day all-cause mortality.

RESULTS

A total of 362 patients were recruited and 319 patients were finally analyzed. Twenty-three patients died during hospitalization and the fatality rate was 7.2%. Furthermore, 14 patients died by the 90-day follow-up. Compared with in-hospital survivors, patients who died in hospital were older (80.78 ± 6.58 vs. 75.09 ± 9.73 years old, P = 0.001), had a higher prevalence of congestive heart failure(69.6% vs. 27.4%, P < 0.001), had a higher BUN/ALB ratio [0.329 (0.250-0.399) vs. 0.145 (0.111-0.210), P < 0.001], had higher neutrophil counts [10.27 (7.21-14.04) vs. 6.58 (4.58-9.04), P < 0.001], higher blood urea nitrogen levels [10.86 (7.10-12.25) vs. 5.35 (4.14-7.40), P < 0.001], a lower albumin level (32.58 ± 3.72 vs. 36.26 ± 4.53, P < 0.001) and a lower lymphocyte count [0.85 (0.58-1.21) vs. 1.22 (0.86-1.72), P = 0.001]. The ROC curve showed that the area under the curve (AUC) of BUN/ALB ratio for in-hospital death was 0.87, (95%CI 0.81-0.93, P < 0.001), the best cut-off point value to discriminate survivors from non-survivors in hospital was 0.249, the sensitivity was 78.3%, the specificity was 86.5%, and Youden's index was 0.648. Having a BUN/ALB ratio ≥ 0.249 was an independent risk factor for both in-hospital and 90-day all-cause mortality after adjustment for relative risk (RR; RR = 15.08, 95% CI 3.80-59.78, P < 0.001 for a multivariate logistic regression analysis) and hazard ratio (HR; HR = 5.34, 95% CI 1.62-17.57, P = 0.006 for a multivariate Cox regression analysis).

CONCLUSION

An elevated BUN/ALB ratio was a strong and independent predictor of in-hospital and 90-day all-cause mortality in patients with AECOPD.

Association of nitrogen balance trajectories with clinical outcomes in critically ill COVID-19 patients: A retrospective cohort study

BACKGROUND & AIMS

The intensity and duration of the catabolic phase in COVID-19 patients can differ between survivors and non-survivors. The purpose of the study was to assess the determinants of, and association between, nitrogen balance trajectories and outcome in critically ill COVID-19 patients.

METHODS

This retrospective monocentric observational study involved patients admitted to the intensive care unit (ICU) of the University Hospital of Clermont Ferrand, France, from January 2020 to May 2021 for COVID-19 pneumonia. Patients were excluded if referred from another ICU, if their ICU length of stay was <72 h, or if they were treated with renal replacement therapy during the first seven days after ICU admission. Data were collected prospectively at admission and during ICU stay. Death was recorded at the end of ICU stay. Comparisons of the time course of nitrogen balance according to outcome were analyzed using two-way ANOVA. At days 3, 5, 7, 10 and 14, uni- and multivariate logistic regression analyses were performed to assess the impact of a non-negative nitrogen-balance on ICU death. To investigate the relationships between nitrogen balance, inflammatory markers and protein intake, linear and non-nonlinear models were run at days 3, 5 and 7, and the amount of protein intake necessary to reach a neutral nitrogen balance was calculated. Subgroup analyses were carried out according to BMI, age, and sex.

RESULTS

99 patients were included. At day 3, a similar negative nitrogen balance was observed in survivors and non-survivors: -16.4 g/d [-26.5, -3.3] and -17.3 g/d [-22.2, -3.8] (p = 0.54). The trajectories of nitrogen balance over time thus differed between survivors and non-survivors (p = 0.01). In survivors, nitrogen balance increased over time, but decreased from day 2 to day 6 in non-survivors, and thereafter increased slowly up to day 14. At days 5 and 7, a non-negative nitrogen-balance was protective from death. Administering higher protein amounts was associated with higher nitrogen balance.

CONCLUSION

We report a prolonged catabolic state in COVID patients that seemed more pronounced in non-survivors than in survivors. Our study underlines the need for monitoring urinary nitrogen excretion to guide the amount of protein intake required by COVID-19 patients.

Association Between Blood Urea Nitrogen Levels and Length of Stay in Patients with Pneumonic Chronic Obstructive Pulmonary Disease Exacerbation: A Secondary Analysis Based on a Multicentre, Retrospective Cohort Study

Purpose

High blood urea nitrogen (BUN) is associated with an elevated risk of mortality in various diseases, such as heart failure and pneumonia. Heart failure and pneumonia are common comorbidities of chronic obstructive pulmonary disease (COPD) exacerbation. However, data on the relationship of BUN levels with length of stay (LOS) in patients with pneumonic COPD exacerbation are sparse. The purpose of this study was to evaluate the correlation between BUN levels and LOS in a cohort of patients with pneumonic COPD exacerbation.

Patients and Methods

The present study was a multicentre, retrospective cohort study. A total of 1226 patients with pneumonic COPD exacerbation were included through a validated algorithm derived from the 10th revision of the International Classification of Diseases and Related Health Problems (ICD-10). It should be noted that the entire study was completed by Shiroshita et al, who uploaded the data to the DATADRYAD website. The author only used these data for secondary analysis.

Results

After adjusting for potential confounders (age, gender), a nonlinear relationship was detected between BUN levels less than 40 mg/dl and LOS. The effect sizes and the confidence intervals on the left and right sides of the inflection point were 0.27 (0.16, 0.39) and −0.17 (−0.34, 0.01), respectively.

Conclusion

High levels of BUN in the hospital may be associated with increased LOS. BUN was positively related to LOS when BUN was less than 40 mg/dl.

Association Between Pre-operative BUN and Post-operative 30-Day Mortality in Patients Undergoing Craniotomy for Tumors: Data From the ACS NSQIP Database

Objective

There is limited evidence to clarify the specific relationship between pre-operative blood urea nitrogen (BUN) and post-operative 30-day mortality in patients undergoing craniotomy for tumors. Therefore, we aimed to investigate this relationship in detail.

Methods

Electronic medical records of 18,642 patients undergoing craniotomy for tumors in the ACS NSQIP from 2012 to 2015 were subjected to secondary retrospective analysis. The principal exposure was pre-operative BUN. Outcome measures were post-operative 30-day mortality. We used binary logistic regression modeling to evaluate the association between them and conducted a generalized additive model and smooth curve fitting (penalized spline method) to explore the potential relationship and its explicit curve shape. We also conducted sensitivity analyses to ensure the robustness of the results and performed subgroup analyses.

Results

A total of 16,876 patients were included in this analysis. Of these, 47.48% of patients were men. The post-operative 30-day mortality of the included cases was 2.49% (420/16,876), and the mean BUN was 16.874 ± 6.648 mg/dl. After adjusting covariates, the results showed that pre-operative BUN was positively associated with post-operative 30-day mortality (OR = 1.020, 95% CI: 1.004, 1.036). There was also a non-linear relationship between BUN and post-operative 30-day mortality, and the inflection point of the BUN was 9.804. For patients with BUN < 9.804 mg/dl, a 1 unit decrease in BUN was related to a 16.8% increase in the risk of post-operative 30-day mortality (OR = 0.832, 95% CI: 0.737, 0.941); for patients with BUN > 9.804 mg/dl, a 1 unit increase in BUN was related to a 2.8% increase in the risk of post-operative 30-day mortality (OR = 1.028, 95% CI: 1.011, 1.045). The sensitivity analysis proved that the results were robust. The subgroup analysis revealed that all listed subgroups did not affect the relationship between pre-operative BUN and post-operative 30-day mortality (P > 0.05).

Conclusion

Our study demonstrated that pre-operative BUN (mg/dl) has specific linear and non-linear relationships with post-operative 30-day mortality in patients over 18 years of age who underwent craniotomy for tumors. Proper pre-operative management of BUN and maintenance of BUN near the inflection point (9.804 mg/dl) could reduce the risk of post-operative 30-day mortality in these cases.

Coexisting Morbidity Burden in Hospitalized Elderly Patients with New-Onset Heart Failure vs Acutely Decompensated Chronic Heart Failure

Coexisting morbidities (CM) are common in patients with heart failure (HF). This study evaluated the CM burden and its clinical significance in elderly hospitalized patients with new-onset (De-novo) HF (n = 84) and acutely decompensated chronic HF (ADCHF) (n = 122). All had HF symptoms associated with: (a) LVEF <50%, or, (b) left ventricular ejection fraction (LVEF) ≥50% and NT-proBNP ≥300 pg/mL. The primary endpoint was the composite of all-cause death/HF rehospitalization at 6 months. Age was similar between patients with new-onset HF and ADCHF [82 (12.5) vs 80 (11) years, respectively; P = .549]. The CM burden was high in both groups. However, the number of CM [3 (2) vs 4 (1.75)] and the prevalence of multimorbidity [CM ≥2; 65 (77.4%) vs 108 (88.5%)] were lower in new-onset HF ( P = .016 and P = .035, respectively). The survival probability without the primary endpoint was higher in new-onset HF than in ADCHF ( P = .001) driven by less rehospitalizations ( P = .001). In the total study population significant primary endpoint predictors were red blood cell distribution width (RDW), urea, and coronary artery disease (CAD) prevalence (AUC of the model =.7685), whereas significant death predictors were RDW, urea, and the number of CM (AUC = .7859), all higher in ADCHF. Thus, the higher CM burden in ADCHF than in new-onset HF most likely contributed to the worse outcome.

Renal Manifestations and their Association with Mortality and Length of Stay in COVID-19 Patients at a Safety-net Hospital

Background

Renal involvement in COVID-19 leads to severe disease and higher mortality. We study renal parameters in COVID-19 patients and their association with mortality and length of stay in hospital.

Methods

A retrospective study (n=340) of confirmed COVID-19 patients with renal involvement determined by the presence of acute kidney injury. Multivariate analyses of logistic regression for mortality and linear regression for length of stay (LOS) adjusted for relevant demographic, comorbidity, disease severity, and treatment covariates.

Results

Mortality was 54.4% and mean LOS was 12.9 days. For mortality, creatinine peak (OR:35.27, 95% CI:2.81, 442.06, p<0.01) and persistent renal involvement at discharge (OR:4.47, 95% CI:1.99,10.06, p<0.001) were each significantly associated with increased odds for mortality. Increased blood urea nitrogen peak (OR:0.98, 95%CI:0.97,0.996, p<0.05) was significantly associated with decreased odds for mortality. For LOS, increased blood urea nitrogen peak (B:0.001, SE:<0.001, p<0.01), renal replacement therapy (B:0.19, SE:0.06, p<0.01), and increased days to acute kidney injury (B:0.19, SE:0.05, p<0.001) were each significantly associated with increased length of stay.

Conclusion

Our study emphasizes the importance in identifying renal involvement parameters in COVID-19 patients. These parameters are associated with LOS and mortality, and may assist clinicians to prognosticate COVID-19 patients with renal involvement.

Recurrent acute kidney injury in elderly patients is common and associated with 1-year mortality

BACKGROUND

Acute kidney injury (AKI) is common among elderly patients after a first hospitalized AKI. Patients who recover are at risk for recurrence, but recurrent geriatric AKI is not well-studied.

METHODS

This was a retrospective, 12-month cohort study using data from the National Clinical Research Center for Geriatric Diseases. Recurrent AKI was defined as a new spontaneous rise of ≥ 0.3 mg/dl (≥ 26.5 µmol/L) within 48 h or a 50% increase in serum creatinine (Scr) from the baseline within 7 days after the previous AKI episode. The outcome measured was 12-month mortality.

RESULTS

Among 1711 study patients, 652 developed AKI. Of the 429 AKI survivors in whom recovery could be assessed, 314 patients recovered to their baseline renal function, and 115 patients developed chronic kidney disease (CKD). Of the group that recovered renal function, 90 patients (28.7%) subsequently developed recurrent AKI, while 224 (71.3%) did not. Of the 429 survivors with AKI, 103 patients (24.0%) died within 12 months. Multivariate logistic regression analysis revealed that recurrent AKI was significantly associated with coronary disease (odds ratio [OR = 2.008; 95% confidence interval [CI] 1.024-3.938; P = 0.042), a need for mechanical ventilation (OR = 2.265; 95% CI 1.267-4.051; P = 0.006) and high blood urea nitrogen levels (OR = 1.036; 95% CI 1.002-1.072; P = 0.040) at the first AKI event. Kaplan-Meier curves showed the 12-month survival of patients with non-recurrent AKI was better than that of patients with CKD, and survival of patients with recurrent AKI was worse than that of patients with CKD (log rank P < 0.001). In the multivariate Cox regression analysis, mortality at 12 month was higher in the patient with recurrent AKI as compared with those with a single episode (HR = 3.375; 95% CI 2.241-5.083; P < 0.001).

CONCLUSION

Recurrent AKI is common among elderly patients who recovered their renal function post-AKI and is associated with significantly higher 12-month mortality compared with CKD patients.

Blood Urea Nitrogen as a Prognostic Marker in Severe Acute Pancreatitis

Objectives

To explore independent risk factors with good and early predictive power for SAP severity and prognosis.

Methods

Patients with SAP were enrolled at Central South University Xiangya Hospital between April 2017 and May 2021 and used as the training cohort. From June 2021 to February 2022, all patients with SAP were defined as external patients for validation. Patients were grouped by survival status at a 30-day posthospital admission and then compared in terms of basic information and laboratory tests to screen the independent risk factors.

Results

A total of 249 patients with SAP were enrolled in the training cohort. The all-cause mortality rate at a 30-day postadmission was 25.8% (51/198). Blood urea nitrogen (BUN) levels were significantly higher in the mortality group (20.45 [interquartile range (IQR), 19.7] mmol/L) than in the survival group (6.685 [IQR, 6.3] mmol/L; P < 0.001). After propensity score matching (PSM), the BUN level was still higher in the mortality group than in the survival group (18.415 [IQR, 19.555] mmol/L vs. 10.63 [IQR, 6.03] mmol/L; P = 0.005). The area under the curve (AUC) of the receiver operating characteristic curve (ROC) of BUN was 0.820 (95% confidence interval, 0.721–0.870; P < 0.001). The optimal BUN level cut-off for predicting a 30-day all-cause mortality was 10.745 mmol/L. Moreover, patients with SAP were grouped according to BUN levels and stratified according to optimal cut-off value. Patients with high BNU levels were associated with significantly higher rates of invasive mechanical ventilation (before PSM: 61.8% vs. 20.6%, P < 0.001; after PSM: 71.1% vs. 32%, P = 0.048) and a 30-day all-cause mortality (before PSM: 44.9% vs. 6.9%, P < 0.001; after PSM: 60% vs. 34.5%, P = 0.032) than those with low BNU levels before or after PSM. The effectiveness of BUN as a prognostic marker was further validated using ROC curves for the external validation set (n = 49). The AUC of BUN was 0.803 (95% CI, 0.655–0.950; P = 0.011). It showed a good ability to predict a 30-day all-cause mortality in patients with SAP. We also observed similar results regarding disease severity, including the Acute Physiology and Chronic Health Evaluation II score (before PSM: 16 [IQR, 8] vs. 8 [IQR, 6], P < 0.001; after PSM: 18 [IQR, 10] vs. 12 [IQR, 7], P < 0.001), SOFA score (before PSM: 7 [IQR, 5] vs. 3 [IQR, 3], P < 0.001; after PSM: 8 [IQR, 5] vs. 5 [IQR, 3.5], P < 0.001), and mMarshall score (before PSM: 4 [IQR, 3] vs. 3 [IQR, 1], P < 0.001; after PSM: 5 [IQR, 2.5] vs. 3 [IQR, 1], P < 0.001). There was significant increase in intensive care unit occupancy in the high BUN level group before PSM (93.3% vs. 73.1%, P < 0.001), but not after PSM (97.8% vs. 86.2%, P = 0.074).

Conclusions

Our results showed that BUN levels within 24 h after hospital admission were independent risk factors for a 30-day all-cause death in patients with SAP.

Prognostic value of albumin-related ratios in HBV-associated decompensated cirrhosis

Background

Identification of effective and accurate prognostic biomarkers for hepatitis B virus‐associated decompensated cirrhosis (HBV‐DeCi) is challenging. This study was designed to determine and compare the prognostic value of albumin‐related ratios (blood urea nitrogen‐to‐albumin ratio [BAR], C‐reactive protein‐to‐albumin ratio [CAR], prothrombin time‐international normalized ratio‐to‐albumin ratio [PTAR], neutrophil count‐to‐albumin ratio [NAR], and D‐dimer‐to‐albumin ratio [DAR]) in HBV‐DeCi patients.

Methods

We retrospectively recruited 161 HBV‐DeCi patients. Receiver operating characteristic curve, DeLong test, and Cox regression analyses were used to estimate and compare the predictive value of these five albumin‐related ratios and Model for End‐Stage Liver Disease (MELD) score.

Results

A total of 29 (18.0%) patients had died 30 days after admission. The prognostic roles of CAR, DAR, PTAR, NAR, and BAR in HBV‐DeCi were different. CAR, PTAR, NAR, and BAR were significantly higher in non‐survivors compared with survivors. However, DAR did not differ between the two groups. The predictive power of BAR was superior to that of the other four albumin‐related biomarkers and similar to that of MELD score. On multivariate analysis, BAR and MELD score were identified as independent prognostic factors, and the combination of BAR and MELD score may improve the prognostic accuracy in HBV‐DeCi.

Conclusion

The present findings suggest that BAR may be a simple and useful prognostic tool to predict mortality in HBV‐DeCi patients.

Biomarkers Predicting Tissue Pharmacokinetics of Antimicrobials in Sepsis: A Review

The pathophysiology of sepsis alters drug pharmacokinetics, resulting in inadequate drug exposure and target-site concentration. Suboptimal exposure leads to treatment failure and the development of antimicrobial resistance. Therefore, we seek to optimize antimicrobial therapy in sepsis by selecting the right drug and the correct dosage. A prerequisite for achieving this goal is characterization and understanding of the mechanisms of pharmacokinetic alterations. However, most infections take place not in blood but in different body compartments. Since tissue pharmacokinetic assessment is not feasible in daily practice, we need to tailor antibiotic treatment according to the specific patient’s pathophysiological processes. The complex pathophysiology of sepsis and the ineffectiveness of current targeted therapies suggest that treatments guided by biomarkers predicting target-site concentration could provide a new therapeutic strategy. Inflammation, endothelial and coagulation activation markers, and blood flow parameters might be indicators of impaired tissue distribution. Moreover, hepatic and renal dysfunction biomarkers can predict not only drug metabolism and clearance but also drug distribution. Identification of the right biomarkers can direct drug dosing and provide timely feedback on its effectiveness. Therefore, this might decrease antibiotic resistance and the mortality of critically ill patients. This article fills the literature gap by characterizing patient biomarkers that might be used to predict unbound plasma-to-tissue drug distribution in critically ill patients. Although all biomarkers must be clinically evaluated with the ultimate goal of combining them in a clinically feasible scoring system, we support the concept that the appropriate biomarkers could be used to direct targeted antibiotic dosing.

Composite Biomarkers for Assessing Frailty Status in Stable Older Adults With Cardiovascular Disease

Background: The relationship between frailty status and laboratory measurements in cardiovascular disease (CVD) remains unclear. We investigated which laboratory measurements indicated frailty in stable older CVD patients.

Methods and Results: One-hundred thirty-eight stable older CVD patients were evaluated by laboratory measurements, with frailty assessed using the Kihon Checklist (KCL). Laboratory measurements were compared between frail and non-frail groups. Across the entire cohort, mean age was 81.7 years, mean left ventricular ejection fraction was 57.8%, and mean plasma B-type natriuretic peptide was 182 pg/mL. KCL scores were used to divide patients into non-frail (n=43; KCL <8) and frail (n=95; KCL ≥8) groups. Serum iron was significantly lower in the frail than non-frail group (mean [±SD] 61.2±30.3 vs. 89.5±26.1 μg/dL, respectively; P<0.001). Blood urea nitrogen (BUN; 27.3±16.5 vs. 19.7±8.2 mg/dL; P=0.013) and C-reactive protein (CRP; 1.05±1.99 vs. 0.15±0.21 mg/dL; P=0.004) were significantly higher in the frail than non-frail group. Multivariate analysis revealed that serum iron, CRP, and BUN were significant independent predictors of frailty (β=−0.069, 0.917, and 0.086, respectively).

Conclusions: Frailty status was significantly associated with iron, CRP, and BUN in stable older CVD patients. Composite biomarkers (inflammation, iron deficiency, and renal perfusion) may be useful for assessing frailty in these patients.

A Novel Approach for COVID-19 Patient Condition Tracking: From Instant Prediction to Regular Monitoring

Purpose: The aim of this research is to develop an accurate and interpretable aggregated score not only for hospitalization outcome prediction (death/discharge) but also for the daily assessment of the COVID-19 patient's condition. Patients and Methods: In this single-center cohort study, real-world data collected within the first two waves of the COVID-19 pandemic was used (27.04.2020-03.08.2020 and 01.11.2020-19.01.2021, respectively). The first wave data (1,349 cases) was used as a training set for the score development, while the second wave data (1,453 cases) was used as a validation set. No overlapping cases were presented in the study. For all the available patients' features, we tested their association with an outcome. Significant features were taken for further analysis, and their partial sensitivity, specificity, and promptness were estimated. Sensitivity and specificity were further combined into a feature informativeness index. The developed score was derived as a weighted sum of nine features that showed the best trade-off between informativeness and promptness. Results: Based on the training cohort (median age ± median absolute deviation 58 ± 13.3, females 55.7%), the following resulting score was derived: APTT (4 points), CRP (3 points), D-dimer (4 points), glucose (4 points), hemoglobin (3 points), lymphocytes (3 points), total protein (6 points), urea (5 points), and WBC (4 points). Internal and temporal validation based on the second wave cohort (age 60 ± 14.8, females 51.8%) showed that a sensitivity and a specificity over 90% may be achieved with an expected prediction range of more than 7 days. Moreover, we demonstrated high robustness of the score to the varying peculiarities of the pandemic. Conclusions: An extensive application of the score during the pandemic showed its potential for optimization of patient management as well as improvement of medical staff attentiveness in a high workload stress. The transparent structure of the score, as well as tractable cutoff bounds, simplified its implementation into clinical practice. High cumulative informativeness of the nine score components suggests that these are the indicators that need to be monitored regularly during the follow-up of a patient with COVID-19.

Role of blood urea nitrogen and serum albumin ratio in predicting severity of community acquired pneumonia (CAP)

Blood urea nitrogen and serum albumin levels are independent risk factors for poor clinical outcome in CAP. However, there is a paucity in the literature on the role of Blood urea nitrogen and albumin ratio(B/A) in CAP. This was a prospective observational study in which 112 admitted patients with the diagnosis of CAP underwent routine blood examinations, ABG, procalcitonin and Chest X-ray. Univariate analysis among various risk factors, CURB-65 scores, blood parameters including B/A ratios and clinical outcomes were carried out followed by multiple logistic regression. Cox regression was done to look at B/A values and time to mortality. In the logistic regression, age, CURB -65 score, B/A ratio and procalcitonin came out to be independent risk factors for ICU admission and mortality. Odds ratio of B/A in predicting mortality and ICU admission came out to be 67.8 (49.2-95.4) and 11.2 (8.4-14), respectively. Cox regression showed B/A values were also found to have a statistically significant relationship with time to mortality (p=0.001). B/A ratio has the potential to become a veritable predictor of poor clinical outcomes in patients with CAP.

The association of blood urea nitrogen levels upon emergency admission with mortality in acute exacerbation of chronic obstructive pulmonary disease

Background and purpose

High blood urea nitrogen (BUN) is associated with an elevated risk of mortality in various diseases, such as heart failure and pneumonia. Heart failure and pneumonia are common comorbidities of acute exacerbation of chronic obstructive pulmonary disease (AECOPD). However, data on the relationship of BUN levels with mortality in patients with AECOPD are sparse. The purpose of this study was to evaluate the correlation between BUN level and in-hospital mortality in a cohort of patients with AECOPD who presented at the emergency department (ED).

Methods

A total of 842 patients with AECOPD were enrolled in the retrospective observational study from January 2018 to September 2020. The outcome was all-cause in-hospital mortality. Receiver operating characteristic (ROC) curve analysis and logistic regression models were performed to evaluate the association of BUN levels with in-hospital mortality in patients with AECOPD. Propensity score matching was used to assemble a cohort of patients with similar baseline characteristics, and logistic regression models were also performed in the propensity score matching cohort.

Results

During hospitalization, 26 patients (3.09%) died from all causes, 142 patients (16.86%) needed invasive ventilation, and 190 patients (22.57%) were admitted to the ICU. The mean level of blood urea nitrogen was 7.5 ± 4.5 mmol/L. Patients in the hospital non-survivor group had higher BUN levels (13.48 ± 9.62 mmol/L vs. 7.35 ± 4.14 mmol/L, p < 0.001) than those in the survivor group. The area under the curve (AUC) was 0.76 (95% CI 0.73–0.79, p < 0.001), and the optimal BUN level cutoff was 7.63 mmol/L for hospital mortality. As a continuous variable, BUN level was associated with hospital mortality after adjusting respiratory rate, level of consciousness, pH, PCO_2, lactic acid, albumin, glucose, CRP, hemoglobin, platelet distribution width, D-dimer, and pro-B-type natriuretic peptide (OR 1.10, 95% CI 1.03–1.17, p=0.005). The OR of hospital mortality was significantly higher in the BUN level ≥7.63 mmol/L group than in the BUN level <7.63 mmol/L group in adjusted model (OR 3.29, 95% CI 1.05–10.29, p=0.041). Similar results were found after multiple imputation and in the propensity score matching cohort.

Conclusions

Increased BUN level at ED admission is associated with hospital mortality in patients with AECOPD who present at the ED. The level of 7.63 mmol/L can be used as a cutoff value for critical stratification.

Predicting 1-Year Mortality on Admission Using the Mayo Cardiac Intensive Care Unit Admission Risk Score

OBJECTIVE

To determine whether the Mayo Cardiac Intensive Care Unit (CICU) Admission Risk Score (M-CARS) accurately predicts 1-year mortality.

METHODS

We retrospectively reviewed adult CICU patients admitted from January 1, 2007, through April 30, 2018, and calculated M-CARS using admission data. We examined the association between admission M-CARS, as continuous and categorical variables, and 1-year mortality.

RESULTS

This study included 12,428 unique patients with a mean age of 67.6±15.2 years (4686 [37.7%] female). A total of 2839 patients (22.8%) died within 1 year of admission, including 1149 (9.2%) hospital deaths and 1690 (15.0%) of the 11,279 hospital survivors. The 1-year survival decreased incrementally as a function of increasing M-CARS (P<.001), and all components of M-CARS were significant predictors of 1-year mortality (P<.001). The 1-year survival among hospital survivors decreased incrementally as a function of increasing M-CARS for scores below 3 (all P<.001); however, there was no further decrease in 1-year survival for hospital survivors with M-CARS of 3 or more (P=.99). The M-CARS components associated with 1-year mortality among hospital survivors included blood urea nitrogen, red blood cell distribution width, Braden skin score, and respiratory failure (all P<.001).

CONCLUSION

M-CARS predicted 1-year mortality among CICU admissions, with a plateau effect at high M-CARS of 3 or more for hospital survivors. Significant added predictors of 1-year mortality among hospital survivors included markers of frailty and chronic illness.

Effect of intermittent or continuous feeding and amino acid concentration on urea-to-creatinine ratio in critical illness

BACKGROUND

We sought to determine whether peaks in essential amino acid (EAA) concentration associated with intermittent feeding may provide anabolic advantages when compared with continuous feeding regimens in critical care.

METHODS

We performed a secondary analysis of data from a multicenter trial of UK intensive care patients randomly assigned to intermittent or continuous feeding. A linear mixed-effects model was developed to assess differences in urea-creatinine ratio (raised values of which can be a marker of muscle wasting) between arms. To investigate metabolic phenotypes, we performed k-means urea-to-creatinine ratio trajectory clustering. Amino acid concentrations were also modeled against urea-to-creatinine ratio from day 1 to day 7. The main outcome measure was serum urea-to-creatinine ratio (millimole per millimole) from day 0 to the end of the 10-day study period.

RESULTS

Urea-to-creatinine ratio trajectory differed between feeding regimens (coefficient -.245; P = .002). Patients receiving intermittent feeding demonstrated a flatter urea-to-creatinine ratio trajectory. With k-means analysis, the cluster with the largest proportion of continuously fed patients demonstrated the steepest rise in urea-to-creatinine ratio. Neither protein intake per se nor serum concentrations of EAA concentrations were correlated with urea-to-creatinine ratio (coefficient = .088 [P = .506] and coefficient <.001 [P = .122], respectively).

CONCLUSION

Intermittent feeding can mitigate the rise in urea-to-creatinine ratio otherwise seen in those continuously fed, suggesting that catabolism may have been, to some degree, prevented.

Comparison of IVC diameter ratio, BUN/creatinine ratio and BUN/albumin ratio for risk prediction in emergency department patients

INTRODUCTION

Early prediction of patients' prognosis in the emergency department (ED) is important. Patients' conditions such as dehydration help predict prognosis. The ratio of serum blood urea nitrogen to creatinine (BUN/Cr ratio) and inferior vena cava (IVC) diameter is often used to determine dehydration. Also, serum albumin levels reflect nutritional conditions such as dehydration. In this study, we evaluated the performance of BUN/Cr ratio, IVC diameter ratio, and BUN/Albumin ratio as predictive markers for in-hospital mortality and ICU admission among various diseases in ED.

MATERIAL AND METHODS

This retrospective cohort study utilized data from patients who had abdominal and pelvic computed tomography (APCT) performed at our institution from 2015 to 2018. The measurement of IVC diameter from computed tomography, the BUN/Cr ratio, and the BUN/Albumin ratio were calculated. Differences in the performance among the BUN/Cr ratio, the IVC diameter ratio, and the BUN/Albumin ratio for predicting outcomes were evaluated by the area under the receiver operating characteristic (AUROC) curve.

RESULTS

A total of 914 patients were enrolled and 78 patients (8.5%) were admitted to the ICU, and 71 patients (7.8%) died during the clinical process. Multivariate logistic regression showed that only the BUN/Albumin ratio was a significant predictor of inhospital mortality and ICU admission.

CONCLUSION

Among dehydration markers the BUN/Albumin ratio is a simple and useful tool for predicting the outcomes of patients visiting the ED.

Urea to albumin ratio is a predictor of mortality in patients with septic shock

BACKGROUND

The purpose of this study was to evaluate urea to albumin ratio (UAR) as predictor of mortality in patients with septic shock.

METHODS

We included all individuals aged ≥ 18 years, with the diagnosis of septic shock at Intensive Care Unit (ICU) admission. Laboratorial and clinical data was recorded within the first 24 h of the patient's admission. Serum urea and albumin concentration were used for UAR calculation. All patients were followed during their ICU stay and the mortality rate was recorded.

RESULTS

222 patients were included in the analysis; the mean age was 62.3 ± 15.1 years and 66% were male. Mortality rate during the ICU stay was 59.9% and the median UAR was 40.7 (24.5-66.1). The UAR was also higher in patients who died in the ICU and was positively correlated with APACHE II, SOFA score and CRP. The ROC ICU mortality development (AUC: 0.617; CI 95%: 0.541-0.693; p: 0.003) at the cutoff of ≥47.25. Furthermore, UAR values were associated with ICU mortality when adjusted by age, sex and APACHE II (OR: 1.011; CI95%:1.000-1.022; p = 0.043) and when adjusted by lactate (OR: 1.014; CI95%:1.003-1.024; p = 0.009).

CONCLUSIONS

Our data suggest that UAR could play a role as predictor of ICU mortality in patients with septic shock.

A clinical nomogram predicting unplanned intensive care unit admission after hip fracture surgery

BACKGROUND

Despite the improvement of surgical procedures and perioperative management, a portion of patients were still at high risk for intensive care unit admission owing to severe morbidity after hip fracture surgeries. The purpose of this study was to analyze influencing factors and to construct a clinical nomogram to predict unscheduled intensive care unit admission among inpatients after hip fracture surgeries.

METHODS

We enrolled a total of 1,234 hip fracture patients, with 40 unplanned intensive care unit admissions, from January 2011 to December 2018. Demographics, chronic coexisting conditions at admission, laboratory tests, and surgical variables were collected and compared between intensive care unit admission and nonadmission groups using univariate analysis. The optimal lasso model was refined to the whole data set, and multivariate logistic regression was used to assign relative weights. A nomogram incorporating these predictors was constructed to visualize these predictors and their corresponding points of the risk for unplanned intensive care unit admission. The model was validated temporally using an independent data set from January 2019 to December 2019 by receiver operating characteristic area under the curve analysis.

RESULTS

In the development group, we identified age, chronic heart failure, coronary heart disease, chronic obstructive pulmonary disease, Parkinson disease, and serum albumin and creatinine concentration were associated with unscheduled intensive care unit admission using multivariate analysis. The final model had an area under the curve of 0.854 (95% confidence interval, 0.742-0.966). The median calculated odds ratio of intensive care unit admission based on the nomogram was significantly higher for patients in the intensive care unit admission group than in the non-intensive care unit admission group (65.93% vs 0.02%, P < .01). The validation group proved its high predictive power with an area under the curve of 0.96 (95% confidence interval, 0.91-0.99).

CONCLUSION

In this study, we identified several independent factors that may increase the risk for unexpected intensive care unit admission after hip fracture surgery and developed a clinical nomogram based on these variables. Preoperative evaluation using this nomogram might facilitate advanced intensive care unit resource management for high-risk patients whose conditions might easily deteriorate if not closely monitored in general wards after surgeries.

Blood Urea Nitrogen and In-Hospital Mortality in Critically Ill Patients with Cardiogenic Shock: Analysis of the MIMIC-III Database

The association between blood urea nitrogen (BUN) and prognosis has been the focus of recent research. Therefore, the objective of this study was to investigate the association between BUN and hospital mortality in critically ill patients with cardiogenic shock (CS). This was a retrospective cohort study, in which data were obtained from the Medical Information Mart for Intensive Care III V1.4 database. Data from 697 patients with CS were analyzed. Logistic regression and subgroup analyses were used to assess the association between BUN and hospital mortality in patients with CS. The average age of the 697 participants was 71.14 years, and approximately 42.18% were men. In the multivariate logistic regression model, after adjusting for age, sex, diabetes, cardiac arrhythmias, urine output, simplified acute physiology score II, sequential organ failure assessment, creatinine, anion gap, and heart rate, high BUN demonstrated strong associations with increased in-hospital mortality (per standard deviation increase: odds ratio [OR] 1.47, 95% confidence interval [CI] 1.13–1.92). A similar result was observed in BUN tertile groups (BUN 23–37 mg/dL versus 6–22 mg/dL: OR [95% CI], 1.42 [0.86–2.34]; BUN 38–165 mg/dL versus 6–22 mg/dL: OR [95% CI], 1.99 [1.10–3.62]; P trend 0.0272). Subgroup analysis did not reveal any significant interactions among various subgroups, and higher BUN was associated with adverse clinical outcomes in patients with CS.

Factors associated with lethality from pneumonia in an acute care for the elderly unit: a retrospective cohort

Introduction: Pneumonia is the leading cause of death due to infection in the world, mainly affecting the older adult population. Objective: To determine the risk factors associated with lethality at 30 days in patients with pneumonia in an acute care for the elderly unit. Materials and methods: We conducted an observational, analytical retrospective cohort study including 114 patients aged 60 years and older hospitalized in an acute care geriatric unit with a diagnosis of pneumonia. The primary outcome was lethality at 30 days. Bivariate and multivariable log-binomial regression analyzes were performed to explore the relationship between independent variables and the primary outcome. Results: The 30-day lethality was 26.3% and the mean age was 84.45 ± 7.37 years; 54.4% of participants were men. In the multivariable analysis, age ≥90 years (Relative Risk, RR=1.62, 95% CI: 1.05-2.68, p=0.04), multilobar commitment (RR=1.92, 95% CI:1.12-3.32, p=0.02), elevated urea nitrogen (≥22.5 the median; RR=3.93, 95% CI:1.67-9.25, p<0.01), and a score of zero in the Lawton index at admission (RR=3.20, 95% CI:1.05-9.78, p=0.04) were independent predictors of 30-day lethality from pneumonia. Conclusion: In older adults hospitalized for pneumonia in an acute care geriatric unit, advanced age, the presence of multilobar commitment, dependency in functional status, and elevated ureic nitrogen levels were the main predictors of short-term lethality risk.

Efficacy of blood urea nitrogen and the neutrophil-to-lymphocyte ratio as predictors of mortality among elderly patients with genitourinary tract infections: A retrospective multicentre study

OBJECTIVES

To investigate whether initial blood urea nitrogen (BUN) and the neutrophil-to-lymphocyte ratio (NLR) in the emergency department (ED) are associated with mortality in elderly patients with genitourinary tract infections.

METHODS

A total of 541 patients with genitourinary tract infections in 5 EDs between November 2016 and February 2017 were included and retrospectively reviewed. We assessed age, sex, comorbidities, vital signs, and initial laboratory results, including BUN, NLR and the SOFA criteria. The primary outcome was all-cause in-hospital mortality.

RESULTS

The nonsurvivor group included 32 (5.9%) elderly patients, and the mean arterial pressure (MAP), NLR and BUN were significantly higher in this group than in the survivor group (p < 0.001, p = 0.003, p < 0.001). In multivariate analysis, MAP <70 mmHg, NLR ≥23.8 and BUN >28 mg/dl were shown to be independent risk factors for in-hospital mortality (OR 3.62, OR 2.51, OR 2.76: p = 0.002, p = 0.033, p = 0.038, respectively). Additionally, NLR ≥23.8 and BUN >28 were shown to be independent risk factors for mortality in admitted elderly with complicated UTI (p = 0.030, p = 0.035). When BUN and NLR were combined with MAP, the area under the ROC curve (AUROC) value was 0.807 (0.771-0.839) for the prediction of mortality, the sensitivity was 87.5% (95% CI 71.0-96.5), and the specificity was 61.3% (95% CI 56.9-65.5%).

CONCLUSION

The initial BUN and NLR values with the MAP were good predictors associated with all-cause in-hospital mortality among elderly genitourinary tract infections visiting the ED.

Blood urea nitrogen to albumin ratio is a predictor of in-hospital mortality in older emergency department patients

INTRODUCTION

In this study, we aimed to evaluate the first measured blood urea nitrogen (BUN)/albumin ratio in the emergency department (ED) as a predictor of in-hospital mortality in older ED patients.

METHODS

This retrospective observational study was conducted at a university hospital ED. Consecutive patients aged 65 and over who visited the ED in a three-month period were included in the study. The BUN, albumin, creatinine, and estimated glomerular filtration rate (eGFR) of patients were recorded. The primary end point of the study was in-hospital mortality.

RESULTS

A total of 1253 patients were included in the statistical analyses of the study. Non-survivors had increased BUN levels (32.9 (23.3-55.4) vs. 20.2 (15.4-28.3) mg/dL, p < 0.001), decreased albumin levels (3.27 (2.74-3.75) vs. 3.96 (3.52-4.25) g/dL, p < 0.001), and increased BUN/albumin ratios (10.19 (6.56-18.94) vs. 5.21 (3.88-7.72) mg/g, p < 0.001) compared to survivors. An increased BUN/albumin ratio was a powerful predictor of in-hospital mortality with an area under the curve of 0.793 (95% CI: 0.753-0.833). Malignancy (OR: 2.39; 95% CI: 1.59-3.74, p < 0.001), albumin level < 3.5 g/dL (OR: 2.75; 95% CI: 1.74-4.36, p < 0.001), and BUN/albumin ratio > 6.25 (OR: 2.82; 95% CI: 1.22-6.50, p < 0.015) were found to be independent predictors of in-hospital mortality in older ED patients.

CONCLUSION

According to our findings, older patients with a BUN level > 23 mg/dL, an albumin level < 3.5 g/dL, and a BUN/albumin ratio > 6.25 mg/g in the ED have a higher risk of in-hospital mortality. Additionally, the BUN/albumin ratio is a more powerful independent predictor of in-hospital mortality than the BUN level, albumin level, creatinine level, and eGFR in older ED patients.

Development and validation of a machine learning-based prediction model for near-term in-hospital mortality among patients with COVID-19

OBJECTIVES

To develop and validate a model for prediction of near-term in-hospital mortality among patients with COVID-19 by application of a machine learning (ML) algorithm on time-series inpatient data from electronic health records.

METHODS

A cohort comprised of 567 patients with COVID-19 at a large acute care healthcare system between 10 February 2020 and 7 April 2020 observed until either death or discharge. Random forest (RF) model was developed on randomly drawn 70% of the cohort (training set) and its performance was evaluated on the rest of 30% (the test set). The outcome variable was in-hospital mortality within 20-84 hours from the time of prediction. Input features included patients' vital signs, laboratory data and ECG results.

RESULTS

Patients had a median age of 60.2 years (IQR 26.2 years); 54.1% were men. In-hospital mortality rate was 17.0% and overall median time to death was 6.5 days (range 1.3-23.0 days). In the test set, the RF classifier yielded a sensitivity of 87.8% (95% CI: 78.2% to 94.3%), specificity of 60.6% (95% CI: 55.2% to 65.8%), accuracy of 65.5% (95% CI: 60.7% to 70.0%), area under the receiver operating characteristic curve of 85.5% (95% CI: 80.8% to 90.2%) and area under the precision recall curve of 64.4% (95% CI: 53.5% to 75.3%).

CONCLUSIONS

Our ML-based approach can be used to analyse electronic health record data and reliably predict near-term mortality prediction. Using such a model in hospitals could help improve care, thereby better aligning clinical decisions with prognosis in critically ill patients with COVID-19.

Lactate/Albumin Ratio as a Predictor of In-Hospital Mortality in Septic Patients Presenting to the Emergency Department

Background: The aim of this study is to evaluate the prognostic value of the Lactate to Albumin (L/A) ratio compared to that of lactate only in predicting morbidity and mortality in sepsis patients.

Methods: This was a single-center retrospective cohort study. All adult patients above the age of 18 with a diagnosis of sepsis who presented between January 1, 2014 and June 30, 2019 were included. The primary outcome was in-hospital mortality.

Results: A total of 1,381 patients were included, 44% were female. Overall in-hospital mortality was 58.4% with the mortalities of sepsis and septic shock being 45.8 and 67%, respectively. 55.5% of patients were admitted to the intensive care unit. The area under the curve value for lactate was 0.61 (95% CI 0.57–0.65, p < 0.001) and for the L/A ratio was 0.67 (95% CI 0.63–0.70, p < 0.001). The cutoff generated was 1.22 (sensitivity 59%, specificity 62%) for the L/A ratio in all septic patients and 1.47 (sensitivity 60%, specificity 67%) in patients with septic shock. The L/A ratio was a predictor of in-hospital mortality (OR 1.53, CI 1.32–1.78, p < 0.001).

Conclusion: The L/A ratio has better prognostic performance than initial serum lactate for in-hospital mortality in adult septic patients.

Elastic Bandage vs Hypertonic Albumin for Diuretic-Resistant Volume-Overloaded Patients in Intensive Care Unit: A Propensity-Match Study

OBJECTIVE

To compare elastic bandage (EB) vs hypertonic albumin solution administration to increase fluid removal by enhancing loop diuretic efficiency (DE) in patients with volume overload and diuretic resistance.

PATIENTS AND METHODS

In this historic cohort study with propensity matching, we included diuretic-resistant adult (≥18 years) patients with volume overload after fluid resuscitation admitted in the intensive care unit from January 1, 2006, through June 30, 2017. Regression models and propensity matching were used to assess the associations of these interventions with changes in DE and other clinical outcomes.

RESULTS

Of 1147 patients (median age, 66; interquartile range [IQR], 56-76 years; 51% [n=590] men), 384 (33%) received EB and 763 (67%) received hypertonic albumin solution. In adjusted models, EB was significantly associated with higher DE compared with hypertonic albumin solution (odds ratio, 1.37; 95% CI, 1.04 to 1.81; P=.004). After propensity matching of 345 pairs, DE remained significantly different between the 2 groups (median, 2111; IQR, 1092 to 4665 mL for EB vs median, 1829; IQR, 1032 to 3436 mL for hypertonic albumin solution; P=.02). EB, male sex, lower baseline serum urea nitrogen level, lower Charlson Comorbidity Index score, and higher baseline left ventricular ejection fraction were DE determinants. The lowest DE quartile (<1073 mL/40-mg furosemide equivalent) following adjustment for known predictors of mortality remained independently associated with higher 90-day death rate (odds ratio, 1.64; 95% CI, 1.13 to 2.36; P=.009).

CONCLUSION

EB use is associated with greater DE than hypertonic albumin solution during the deescalation phase of sepsis resuscitation. Prospective clinical trials would validate the findings of this hypothesis-generating study.

Evaluation of blood urea nitrogen concentration and anorexia as predictors of nonsurvival in client-owned rabbits evaluated at a veterinary referral center

OBJECTIVE To determine the prognostic relevance of BUN concentration in client-owned rabbits evaluated at a veterinary referral center. ANIMALS 246 healthy or clinically ill client-owned rabbits with BUN concentrations measured at a veterinary referral center. PROCEDURES In a retrospective cohort study design, medical records of rabbits were retrieved, and data were collected on BUN concentration (exposure variable of interest) and other variables, including outcome (survival status at 15 days after BUN concentration measurement). Univariate, multivariate, and subgroup analyses were performed to identify variables associated with outcome. RESULTS BUN concentrations ranged from 6.5 to 251.1 mg/dL (median, 18.7 mg/dL). Univariate analysis revealed that the risk of nonsurvival over the 15-day period for rabbits with a high BUN concentration (≥ 23.3 mg/dL) was 33% higher (relative risk, 1.33; 95% confidence interval [CI], 1.06 to 1.69) than that for rabbits with unremarkable BUN values. Subgroup analysis revealed that for rabbits with anorexia, a high (vs unremarkable) BUN concentration was associated with an increased risk of nonsurvival (relative risk, 1.69; 95% CI, 1.05 to 2.71). In the final multivariate model that controlled for age, sex, and appetite (anorexia vs no anorexia), the odds of nonsurvival for rabbits with BUN values > 24.74 mg/dL were 3 times that for rabbits with BUN values < 14.00 mg/dL (OR, 2.92; 95% CI, 1.29 to 6.58). CONCLUSIONS AND CLINICAL RELEVANCE Results indicated that a high BUN concentration increased the risk of nonsurvival over a 15-day period for client-owned rabbits, particularly those with anorexia. Blood urea nitrogen concentration should be used together with other clinical indicators to provide prognostic information for rabbits receiving veterinary care.