Predicting the need for ICU admission in community-acquired pneumonia

How to assess survival prognosis in patients hospitalized for community-acquired pneumonia in 2024?

PURPOSE OF REVIEW

Community-acquired pneumonia (CAP) is increasingly recognized as a complex, multisystemic disease with the potential to cause both acute and long-term sequelae, significantly impacting patient mortality rates. In this manuscript, the authors review the current methodologies for assessing mortality risk among CAP patients.

RECENT FINDINGS

The most common prediction scores for ICU care and short-term mortality include Pneumonia Severity Index (PSI), CURB-65, SMART COP, SCAP, and ATS/IDSA criteria. These models have clinical utility in the prediction of short-term mortality, but they have significant limitations in addressing long-term mortality. For patients who are discharged alive from the hospital, we do not have scores to predict long term mortality.

SUMMARY

The development of an optimal prognostic tool for postacute sequelae of CAP is imperative. Such a tool should identify specific populations at increased risk. Moreover, accurately identifying at-risk populations is essential for their inclusion in clinical trials that evaluate potential therapies designed to improve short and long-term clinical outcomes in patients with CAP.

An explainable machine learning-based model to predict intensive care unit admission among patients with community-acquired pneumonia and connective tissue disease

BACKGROUND

There is no individualized prediction model for intensive care unit (ICU) admission on patients with community-acquired pneumonia (CAP) and connective tissue disease (CTD) so far. In this study, we aimed to establish a machine learning-based model for predicting the need for ICU admission among those patients.

METHODS

This was a retrospective study on patients admitted into a University Hospital in China between November 2008 and November 2021. Patients were included if they were diagnosed with CAP and CTD during admission and hospitalization. Data related to demographics, CTD types, comorbidities, vital signs and laboratory results during the first 24 h of hospitalization were collected. The baseline variables were screened to identify potential predictors via three methods, including univariate analysis, least absolute shrinkage and selection operator (Lasso) regression and Boruta algorithm. Nine supervised machine learning algorithms were used to build prediction models. We evaluated the performances of differentiation, calibration, and clinical utility of all models to determine the optimal model. The Shapley Additive Explanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) techniques were performed to interpret the optimal model.

RESULTS

The included patients were randomly divided into the training set (1070 patients) and the testing set (459 patients) at a ratio of 70:30. The intersection results of three feature selection approaches yielded 16 predictors. The eXtreme gradient boosting (XGBoost) model achieved the highest area under the receiver operating characteristic curve (AUC) (0.941) and accuracy (0.913) among various models. The calibration curve and decision curve analysis (DCA) both suggested that the XGBoost model outperformed other models. The SHAP summary plots illustrated the top 6 features with the greatest importance, including higher N-terminal pro-B-type natriuretic peptide (NT-proBNP) and C-reactive protein (CRP), lower level of CD4 + T cell, lymphocyte and serum sodium, and positive serum (1,3)-β-D-glucan test (G test).

CONCLUSION

We successfully developed, evaluated and explained a machine learning-based model for predicting ICU admission in patients with CAP and CTD. The XGBoost model could be clinical referenced after external validation and improvement.

Definition, Epidemiology, and Pathogenesis of Severe Community-Acquired Pneumonia

The clinical presentation of community-acquired pneumonia (CAP) can vary widely among patients. While many individuals with mild symptoms can be managed as outpatients with excellent outcomes, there is a distinct subgroup of patients who present with severe CAP. In these cases, the mortality rate can reach approximately 25% within 30 days and even up to 50% within a year. It is crucial to focus attention on these patients who are at higher risk. Among the various definitions of severe CAP found in the literature, one commonly used criterion is the requirement for admission to intensive care unit. Notable epidemiological characteristics of these patients include the impact of acute cardiovascular diseases on clinical outcomes and the enduring, independent effect of pneumonia on long-term outcomes. Factors such as pathogen virulence, the presence of comorbidities, and the host response are important contributors to the pathogenesis of severe CAP. In these patients, the host response may be dysregulated and compartmentalized. Gaining a better understanding of the epidemiology and pathogenesis of severe CAP will provide a foundation for the development of new therapies for this condition. This manuscript aims to review the definition, epidemiology, and pathogenesis of severe CAP, shedding light on important aspects that can aid in the improvement of patient care and outcomes.

Development of a machine learning-based clinical decision support system to predict clinical deterioration in patients visiting the emergency department

This study aimed to develop a machine learning-based clinical decision support system for emergency departments based on the decision-making framework of physicians. We extracted 27 fixed and 93 observation features using data on vital signs, mental status, laboratory results, and electrocardiograms during emergency department stay. Outcomes included intubation, admission to the intensive care unit, inotrope or vasopressor administration, and in-hospital cardiac arrest. eXtreme gradient boosting algorithm was used to learn and predict each outcome. Specificity, sensitivity, precision, F1 score, area under the receiver operating characteristic curve (AUROC), and area under the precision-recall curve were assessed. We analyzed 303,345 patients with 4,787,121 input data, resampled into 24,148,958 1 h-units. The models displayed a discriminative ability to predict outcomes (AUROC > 0.9), and the model with lagging 6 and leading 0 displayed the highest value. The AUROC curve of in-hospital cardiac arrest had the smallest change, with increased lagging for all outcomes. With inotropic use, intubation, and intensive care unit admission, the range of AUROC curve change with the leading 6 was the highest according to different amounts of previous information (lagging). In this study, a human-centered approach to emulate the clinical decision-making process of emergency physicians has been adopted to enhance the use of the system. Machine learning-based clinical decision support systems customized according to clinical situations can help improve the quality of care.

Association of admission lactate with mortality in adult patients with severe community-acquired pneumonia

PURPOSE

The present study was conducted to investigate the association of admission lactate with mortality in severe community-acquired pneumonia (SCAP).

METHODS

We performed a retrospective, observational, cohort study on adult SCAP patients admitted to intensive care unit (ICU) in West China Hospital of Sichuan University between December 2011 and December 2018. The primary outcome was hospital mortality. Univariate and then multivariate analysis were performed to identify independent risk factors for hospital mortality. The association of admission lactate categories with hospital mortality was examined in three logistic regression models and Kaplan-Meier plots. We also applied restricted cubic splines to estimate the potential non-linear associations.

RESULTS

In total, 2275 SCAP patients were included. Admission lactate remained a significant factor for mortality after multivariate regression (OR: 1.085; 95% CI: 1.033,1.141; by continuous variable). After lactate was categorized into quartiles and the confounders were fully adjusted, compared with the quartile 1, ORs (95% CIs) of hospital mortality for quartile 2, quartile 3 and quartile 4 were 1.001 (0.759-1.321), 1.153 (0.877-1.516) and 1.593 (1.202-2.109), respectively (P for trend =0.001). Survival curves indicated that elevated lactate was associated with poor prognosis (P < 0.001). Moreover, this association was non-linear, indicating that increased lactate has the most notable impact on mortality within the range of 1.5 to 4 mmol/L (P non-linear: 0.029 for hospital mortality; 0.004 for ICU mortality).

CONCLUSION

Elevated admission lactate has a significant, independent, and potentially non-linear association with increased mortality in SCAP patients.

IL-7/IL7R axis dysfunction in adults with severe community-acquired pneumonia (CAP): a cross-sectional study

Community-acquired pneumonia (CAP) is a worldwide leading cause of death. Recognized risk factors in some severe cases have not been identified. Lymphocytopenia has been frequently described in CAP. Since IL-7, membrane-bound receptor (IL7Rα;CD127) and soluble IL7Rα (sIL7R) are critical in lymphocytes homeostasis, in this work we aimed to evaluate the involvement of the IL-7/IL7Rα axis in the severity of adult CAP, since it has not been explored. The IL7Rα SNPs rs6897932, rs987106, and rs3194051 SNPs in IL7α were genotyped, the systemic expression of the IL7R gene, sIL7R, IL-7, and levels of peripheral IL7Rα⁺ T lymphocytes were quantified in 202 hospitalized CAP cases. rs3194051GG was more frequent in non-survivors than in survivors; rs987106TT was more frequent and rs3194051AA less frequent in patients at intensive care unit (ICU) than in those not admitted to ICU. IL7Rα gene expression was lower in non-survivors than in survivors, and in severe than in mild cases. CD3⁺CD127⁺ lymphocytes were lower in severe than in mild cases; in non-survivors than in survivors and in ICU than in non- ICU admitted cases. sIL7Rα plasmatic levels were higher in non-survivors than in survivors, and in severe than in mild cases. rs6897932CC, rs987106AA and rs3194051GG carriers showed the highest while rs6897932TT showed the lowest sIL7Rα levels. The AUC of sIL7Rα levels predicting 30-day mortality was 0.71. Plasma IL-7 levels were lower in ICU-admitted than in not ICU-admitted and in non-survivors than in survivors. No additional association was detected. In conclusion, rs3194051GG and rs987106TT IL7R genotypes were associated with a poorer prognosis. A significant association between sIL7R levels and SNPs of the IL7R gene is described for the first time in adult CAP. Increased plasmatic sIL7R could contribute to identifying adult CAP cases at risk of death.

A literature review of severity scores for adults with influenza or community-acquired pneumonia - implications for influenza vaccines and therapeutics

Influenza vaccination and antiviral therapeutics may attenuate disease, decreasing severity of illness in vaccinated and treated persons. Standardized assessment tools, definitions of disease severity, and clinical endpoints would support characterizing the attenuating effects of influenza vaccines and antivirals. We review potential clinical parameters and endpoints that may be useful for ordinal scales evaluating attenuating effects of influenza vaccines and antivirals in hospital-based studies. In studies of influenza and community-acquired pneumonia, common physiologic parameters that predicted outcomes such as mortality, ICU admission, complications, and duration of stay included vital signs (hypotension, tachypnea, fever, hypoxia), laboratory results (blood urea nitrogen, platelets, serum sodium), and radiographic findings of infiltrates or effusions. Ordinal scales based on these parameters may be useful endpoints for evaluating attenuating effects of influenza vaccines and therapeutics. Factors such as clinical and policy relevance, reproducibility, and specificity of measurements should be considered when creating a standardized ordinal scale for assessment.

Cyclovirus detection in Chilean adults with and without community-acquired pneumonia

Cycloviruses (CyV) (genus Cyclovirus, family Circoviridae) are nonenveloped DNA viruses. The first report in humans was in 2010 and research has focused only on disease-associated human sample detection. The only HuACyV (CyCV-ChileNPA1, HuACyV10) reported in the Chilean population was in children (3.3%) with an acute respiratory infection. Its detection in respiratory samples from adults, with/without respiratory disease remains unknown. The aim of this study was to detect HuACyV10 in adults with and without respiratory disease. HuACyV10 was studied in nasopharyngeal swabs from 105 hospitalized adults with community-acquired pneumonia (CAP) and 104 adults without respiratory symptoms. Total nucleic acids were extracted, and viral rep and cp gene fragments were amplified by real-time polymerase chain reaction. HuACyV10 was detected in 19.05% adults with CAP and in 0.96% asymptomatic adults, being significantly higher in adult CAP than asymptomatic (n = 1) ones (p = 0.0001). C _t values were between 26.7 and 39.6, and the median was 34.1 for rep and 33.8 for the CAP in adults CAP (p = 0.68), and 35.7 and 36.0, respectively, in the asymptomatic case. HuACyV10 detection in CAP adults concentrated in the Autumn-Winter season of the Southern hemisphere. The only asymptomatic adult with HuACyV10 was detected in the Spring-Summer period. In this first report of HuACyV10 in respiratory samples from adults, detection was significantly higher in CAP than in asymptomatic adults. As the sensitivity of both rep and cp genes was similar, both can be applied for detecting HuACyV10. It would be advisable to investigate the pathogenic role of HuACyV10 in adult respiratory infections. ​.

Immunoglobulins concentration and B cell counts as severity markers in adult community-acquired pneumonia: Cross sectional study

Community-acquired pneumonia (CAP) is a worldwide cause of morbidity and mortality. Immunoglobulins (Igs) and B cells quantification studies in CAP are few and show discrepancies. Serum IgA acts as a powerful natural anti-inflammatory factor, but its role in the CAP has not yet been defined. The highly sensitive xMAP Luminex technique allows better immunoglobulins quantification. The aim of this study was to analyze the relation between clinical severity and circulating Igs and B cells in adults with CAP.Igs (M, A, G1, G2, G3, and G4) and B cells were quantified in peripheral blood of 190 Chilean patients ≥18 years old hospitalized for CAP and in 21 adults without respiratory disease, using xMAP Luminex and flow cytometry, respectively. Clinical history was recorded and PSI and CURB-65 scores were calculated for evaluation of clinical severity.The total IgM, IgG2 and total IgG levels were lower in CAP than in asymptomatic adults (P < .05). No significant differences of Igs levels were found between patients classified as severe and mild by PSI and CURB-65 scores. Fatal cases had higher levels of IgA (P < .05). No differences in CD19 B cells frequency was found between CAP and asymptomatic adults (P = .40). In PSI severe cases, CD19 B cells were significantly lower than in mild cases (P = .008). No differences were found in CURB-65 severe and mild groups (P = .11). In fatal cases (11/82) group, CD19 B cells frequency was lower than in 71 survivors (P = .2). No differences in memory B lymphocytes were detected between asymptomatic and CAP adults, severe and mild patients, survivors and fatal cases (P > .05).Serum IgA levels were significantly higher in fatal CAP cases, raising it as a potential biomarker for severe disease considering its relatively universal availability. In PSI severe patients, B cells showed lower levels and could have a role on its physiopathology. Finding new markers rooted in physiopathology could improve the possibility of scoring severe CAP cases. Luminex technology showed promising quantification serum Igs.

Comorbidity and its impact on 1590 patients with COVID-19 in China: a nationwide analysis

Background

The coronavirus disease 2019 (COVID-19) outbreak is evolving rapidly worldwide.

Objective

To evaluate the risk of serious adverse outcomes in patients with COVID-19 by stratifying the comorbidity status.

Methods

We analysed data from 1590 laboratory confirmed hospitalised patients from 575 hospitals in 31 provinces/autonomous regions/provincial municipalities across mainland China between 11 December 2019 and 31 January 2020. We analysed the composite end-points, which consisted of admission to an intensive care unit, invasive ventilation or death. The risk of reaching the composite end-points was compared according to the presence and number of comorbidities.

Results

The mean age was 48.9 years and 686 (42.7%) patients were female. Severe cases accounted for 16.0% of the study population. 131 (8.2%) patients reached the composite end-points. 399 (25.1%) reported having at least one comorbidity. The most prevalent comorbidity was hypertension (16.9%), followed by diabetes (8.2%). 130 (8.2%) patients reported having two or more comorbidities. After adjusting for age and smoking status, COPD (HR (95% CI) 2.681 (1.424–5.048)), diabetes (1.59 (1.03–2.45)), hypertension (1.58 (1.07–2.32)) and malignancy (3.50 (1.60–7.64)) were risk factors of reaching the composite end-points. The hazard ratio (95% CI) was 1.79 (1.16–2.77) among patients with at least one comorbidity and 2.59 (1.61–4.17) among patients with two or more comorbidities.

Conclusion

Among laboratory confirmed cases of COVID-19, patients with any comorbidity yielded poorer clinical outcomes than those without. A greater number of comorbidities also correlated with poorer clinical outcomes.

The presence and number of comorbidities predict clinical outcomes of COVID-19http://bit.ly/3b9ibw5

The Burden of Community-Acquired Pneumonia Requiring Admission to ICU in the United States

BACKGROUND

A paucity of studies have assessed the epidemiology of community-acquired pneumonia (CAP) that require ICU admission. We conducted a study on this group of patients with the primary objective of defining the incidence, epidemiology, and mortality rate of CAP in the ICUs in Louisville, Kentucky. The secondary objective was to estimate the number of patients who were hospitalized and the number of deaths that were associated with CAP in ICU in the United States.

RESEARCH QUESTIONS

What is epidemiology of CAP in the ICU in Louisville, Kentucky, and the projected incidence in the United States?

STUDY DESIGN AND METHODS

This was a secondary analysis of a prospective population-based cohort study. The setting was all nine adult hospitals in Louisville, Kentucky. The annual incidence of CAP in the ICU per 100,000 adults was calculated for the whole adult population of Louisville. The number of patients who were hospitalized because of CAP in ICU in the United States was estimated by multiplying the Louisville incidence rate of CAP in ICU by the 2014 US adult population.

RESULTS

From a total of 7,449 unique patients who were hospitalized with CAP, 1,707 patients (23%) were admitted to the ICU. The incidence of CAP in the ICU was 145 cases per 100,000 population of adults. Cases of CAP in the ICU were clustered in patients from areas of the city with high poverty. The mortality rate of patients with CAP in ICU was 27% at 30 days and 47% at one year. In the United States, the estimated number of patients who were hospitalized with CAP requiring the ICU was 356,326 per year, and the estimated number of deaths at 30 days and one year were 96,206 and 167,474, respectively.

INTERPRETATION

Almost one in five patients who are hospitalized with CAP requires intensive care. Poverty is associated with CAP in the ICU. Nearly one-half of patients with CAP in the ICU will die within one year. Because of its significant burden, CAP in the ICU should be a high priority in research agenda and health policy.

Comorbidity and its impact on 1590 patients with COVID-19 in China: a nationwide analysis

BACKGROUND

The coronavirus disease 2019 (COVID-19) outbreak is evolving rapidly worldwide.

OBJECTIVE

To evaluate the risk of serious adverse outcomes in patients with COVID-19 by stratifying the comorbidity status.

METHODS

We analysed data from 1590 laboratory confirmed hospitalised patients from 575 hospitals in 31 provinces/autonomous regions/provincial municipalities across mainland China between 11 December 2019 and 31 January 2020. We analysed the composite end-points, which consisted of admission to an intensive care unit, invasive ventilation or death. The risk of reaching the composite end-points was compared according to the presence and number of comorbidities.

RESULTS

The mean age was 48.9 years and 686 (42.7%) patients were female. Severe cases accounted for 16.0% of the study population. 131 (8.2%) patients reached the composite end-points. 399 (25.1%) reported having at least one comorbidity. The most prevalent comorbidity was hypertension (16.9%), followed by diabetes (8.2%). 130 (8.2%) patients reported having two or more comorbidities. After adjusting for age and smoking status, COPD (HR (95% CI) 2.681 (1.424-5.048)), diabetes (1.59 (1.03-2.45)), hypertension (1.58 (1.07-2.32)) and malignancy (3.50 (1.60-7.64)) were risk factors of reaching the composite end-points. The hazard ratio (95% CI) was 1.79 (1.16-2.77) among patients with at least one comorbidity and 2.59 (1.61-4.17) among patients with two or more comorbidities.

CONCLUSION

Among laboratory confirmed cases of COVID-19, patients with any comorbidity yielded poorer clinical outcomes than those without. A greater number of comorbidities also correlated with poorer clinical outcomes.