Modelos predictivos de bacteriemia en el servicio de urgencias: revisión sistemática / Models to predict bacteremia in the emergency department: a systematic review

Objetivo. La obtención de hemocultivos (HC) se realiza en el 15% de los pacientes atendidos con sospecha de infección en los servicios de urgencias (SU) con una rentabilidad diagnóstica variable (2-20%). La mortalidad a 30 días de estos pacientes con bacteriemia es elevada, doble o triple que el resto con el mismo proceso. Así, encontrar un modelo predictivo de bacteriemia eficaz y aplicable en los SU sería muy importante. Clásicamente, el modelo de Shapiro ha sido la referencia en todo el mundo. El objetivo de esta revisión sistemática (RS) es comparar la capacidad para predecir bacteriemia en los SU de los distintos modelos predictivos publicados desde el año 2008 (fecha de publicación del modelo de Shapiro). Métodos. Se realiza una RS siguiendo la normativa PRISMA en las bases de datos de PubMed, Web of Science, EMBASE, Lilacs, Cochrane, Epistemonikos, Tripdatabase y ClinicalTrials.gov desde enero de 2008 hasta 31 mayo 2023 sin restricción de idiomas y utilizando una combinación de términos MESH: “Bacteremia/Bacteraemia/Blood Stream Infection”, “Prediction Model/Clinical Prediction Rule/Risk Prediction Model”, “Emergencies/Emergency/Emergency Department” y “Adults”. Se incluyeron estudios de cohortes observacionales (analíticos de rendimiento diagnóstico). Para valorar la calidad del método empleado y el riesgo de sesgos de los artículos incluidos se utilizó la NewcastleOttawa Scale (NOS). No se incluyeron estudios de casos y controles, revisiones narrativas y en otros tipos de artículos. No se realizaron técnicas de metanálisis, pero los resultados se compararon narrativamente. El protocolo de la RS se registró en PROSPERO (CRD42023426327). Resultados. Se identificaron 917 artículos y se analizaron finalmente 20 que cumplían los criterios de inclusión. Los estudios incluidos contienen 33.182 HC procesados con 5.074 bacteriemias (15,3%). Once estudios fueron calificados de calidad alta, 7 moderada y 2 baja... (AU)

Objective. Blood cultures are ordered in emergency departments for 15% of patients with suspected infection. The diagnostic yield varies from 2% to 20%. Thirty-day mortality in patients with bacteremia is high, doubling or tripling the rate in patients with the same infection but without bacteremia. Thus, finding an effective model to predict bacteremia that is applicable in emergency departments is an important goal. Shapiro’s model is the one traditionally used as a reference internationally. The aim of this systematic review was to compare the predictive power of bacteremia risk models published since 2008, when Shapiro’s model first appeared. Methods. We followed the recommendations of the Preferred Reporting Items for Systematic Reviews and MetaAnalyses (PRISMA) statement, searching in the following databases for articles published between January 2008 and May 31, 2023: PubMed, Web of Science, EMBASE, Lilacs, Cochrane, Epistemonikos, Trip Medical Database, and ClinicalTrials.gov. No language restrictions were specified. The search terms were the following Medical Subject Headings: bacteremia/bacteraemia/blood stream infection, prediction model/clinical prediction rule/risk prediction model, emergencies/emergency/emergency department, and adults. Observational cohort studies analyzing diagnostic yield were included; case-control studies, narrative reviews, and other types of articles were excluded. The Newcastle-Ottawa Scale was used to score quality and risk of bias in the included studies. The results were compared descriptively, without meta-analysis. The protocol was included in the PROSPERO register (CRD42023426327). Results. Twenty studies out of a total of 917 were found to meet the inclusion criteria. The included studies together analyzed 33 182 blood cultures, which detected 5074 cases of bacteremia (15.3%). Eleven studies were of high quality, 7 of moderate quality, and 2 of low quality... (AU)

Precisión diagnóstica de la procalcitonina para la bacteriemia en el servicio de urgencias: una revisión sistemática / Diagnostic accuracy of procalcitonin for bacteremia in the emergency department: a systematic review

Introducción. La obtención de hemocultivos (HC) se realiza en el 15% de los pacientes atendidos con sospecha de infección en los Servicios de Urgencias Hospitalarios (SUH) con unarentabilidad diagnóstica variable (2-20%). La mortalidad a 30 días de los pacientes con bacteriemia es dos o tres veces mayor que el resto con el mismo proceso. La procalcitonina (PCT) es un biomarcador que se ha utilizado como herramienta de ayuda en la predicción de bacteriemia en los SUH. El objetivo principal de esta revisión sistemática es investigar la precisión diagnóstica de la PCT para predecir bacteriemia verdadera en los pacientes adultos atendidos con sospecha clínica de infección enel SUH, así como identificar un valor específico de PCT como el más relevante desde el punto de vista diagnóstico de decisión clínica que pueda recomendarse para la toma de decisiones. Método. Se realiza una revisión sistemática siguiendo la normativa PRISMA en las bases de datos de PubMed, Web of Sciencie, EMBASE, Lilacs, Cochrane, Epistemonikos, Tripdatabase y ClinicalTrials.gov desde enero de 2010 hasta 31 mayo 2023 sin restricción de idiomas y utilizando una combinación de términos MESH: “Bacteremia/Bacteraemia/Blood Stream Infection”, “Procalcitonin”, “Emergencies/Emergency/ Emergency Department” y “Adults”. Se incluyeron estudios de cohortes observacionales y parcialmente una revisión sistemática. No se realizaron técnicas de metaanálisis, pero los resultados se compararon narrativamente. Resultados. Se identificaron un total de 1.372 artículos de los cuales se analizaron finalmente 20 que cumplían los criterios de inclusión. Los estudios incluidos representan un total de 18.120 HC procesados con 2.877 bacteriemias (15,88%). Diez estudios fueron calificados de calidad alta, 9 moderada y 1 baja. El ABC-COR de todos los estudios se sitúa desde 0,68 (IC 95%: 0,59-0,77) hasta 0,98 (IC 95%: 0,97-0,99). . ...Conclusiones. ... . (AU)

Introduction. Obtaining blood cultures (HC) is performed in 15% of the patients treated with suspicion of infection in the Hospital Emergency Services (ED) with a variable diagnostic yield (2-20%). The 30-day mortality of patients with bacteremia is two or three times higher than the rest with the same process. Procalcitonin (PCT) is a biomarker that has been used as a tool to help predict bacteremia in HEDs. The main objective of this systematic review is to investigate the diagnostic accuracy of PCT in predicting true bacteraemia in adult patients treated with clinical suspicion of infection in the ED, as well as to identify a specific PCT value as the most relevant from the clinical decision diagnostic point of view that can be recommended for decision making. Method. A systematic review was performed following the PRISMA guidelines in the PubMed, Web of Science, EMBASE, Lilacs, Cochrane, Epistemonikos, Tripdatabase and ClinicalTrials. gov databases from January 2010 to May 31, 2023 without language restrictions and using a combination of MESH terms: “Bacteremia/ Bacteraemia/ Blood Stream Infection”, “Procalcitonin”, “Emergencies/ Emergency/ Emergency Department” and “Adults”. Observational cohort studies and partially an systematic review were included. No meta-analysis techniques were performed, but the results were compared narratively. Results. A total of 1,372 articles were identified, of which 20 that met the inclusion criteria were finally analyzed. The included studies represent a total of 18,120 processed HC with 2,877 bacteraemias (15.88%). Ten studies were rated as high, 9 moderate and 1 low quality. The AUC-COR of all the studies ranges from 0.68 (95% CI: 0.59-0.77) to 0.98 (95% CI: 0.97- 0.99). The PCT value >0.5 ng/ml is the most widely used and proposed in up to ten of the works included in this systematic review, whose estimated mean yield is an AUC-COR of 0.833. ...Conclusions. ... (AU)

Models to predict bacteremia in the emergency department: a systematic review. / Modelos predictivos de bacteriemia en el servicio de urgencias: revisión sistemática.

OBJECTIVES: Blood cultures are ordered in emergency departments for 15% of patients with suspected infection. The diagnostic yield varies from 2% to 20%. Thirty-day mortality in patients with bacteremia is high, doubling or tripling the rate in patients with the same infection but without bacteremia. Thus, finding an effective model to predict bacteremia that is applicable in emergency departments is an important goal. Shapiro's model is the one traditionally used as a reference internationally. The aim of this systematic review was to compare the predictive power of bacteremia risk models published since 2008, when Shapiro's model first appeared. MATERIAL AND METHODS: We followed the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement, searching in the following databases for articles published between January 2008 and May 31, 2023: PubMed, Web of Science, EMBASE, Lilacs, Cochrane, Epistemonikos, Trip Medical Database, and ClinicalTrials.gov. No language restrictions were specified. The search terms were the following Medical Subject Headings: bacteremia/bacteraemia/blood stream infection, prediction model/clinical prediction rule/risk prediction model, emergencies/emergency/emergency department, and adults. Observational cohort studies analyzing diagnostic yield were included; case-control studies, narrative reviews, and other types of articles were excluded. The Newcastle-Ottawa Scale was used to score quality and risk of bias in the included studies. The results were compared descriptively, without meta-analysis. The protocol was included in the PROSPERO register (CRD42023426327). RESULTS: Twenty studies out of a total of 917 were found to meet the inclusion criteria. The included studies together analyzed 33 182 blood cultures, which detected 5074 cases of bacteremia (15.3%). Eleven studies were of high quality, 7 of moderate quality, and 2 of low quality. The area under the receiver operating characteristic curve (AUC) of Shapiro's model varied from 0.71 to 0.83. Sensitivity was as high as 98%, and specificity ranged from 26% to 69%. Three models with high scores for quality were also supported by both internal and external validation studies: Lee's model (AUC, 0.81; sensitivity 68%; specificity, 81%), the 5MPB-Toledo model (AUC, 0.906 to 0.946), and the MPB-INFURG-SEMES model (AUC, 0.924; sensitivity, 97%; specificity, 76%. CONCLUSION: The 5MPB-Toledo and MPB-INFURG-SEMES are useful for assessing the true risk of bacteremia in patients attended in emergency departments.

OBJETIVO: La obtención de hemocultivos (HC) se realiza en el 15% de los pacientes atendidos con sospecha de infección en los servicios de urgencias (SU) con una rentabilidad diagnóstica variable (2-20%). La mortalidad a 30 días de estos pacientes con bacteriemia es elevada, doble o triple que el resto con el mismo proceso. Así, encontrar un modelo predictivo de bacteriemia eficaz y aplicable en los SU sería muy importante. Clásicamente, el modelo de Shapiro ha sido la referencia en todo el mundo. El objetivo de esta revisión sistemática (RS) es comparar la capacidad para predecir bacteriemia en los SU de los distintos modelos predictivos publicados desde el año 2008 (fecha de publicación del modelo de Shapiro). METODO: Se realiza una RS siguiendo la normativa PRISMA en las bases de datos de PubMed, Web of Science, EMBASE, Lilacs, Cochrane, Epistemonikos, Tripdatabase y ClinicalTrials.gov desde enero de 2008 hasta 31 mayo 2023 sin restricción de idiomas y utilizando una combinación de términos MESH: "Bacteremia/Bacteraemia/Blood Stream Infection", "Prediction Model/Clinical Prediction Rule/Risk Prediction Model", "Emergencies/Emergency/Emergency Department" y "Adults". Se incluyeron estudios de cohortes observacionales (analíticos de rendimiento diagnóstico). Para valorar la calidad del método empleado y el riesgo de sesgos de los artículos incluidos se utilizó la Newcastle-Ottawa Scale (NOS). No se incluyeron estudios de casos y controles, revisiones narrativas y en otros tipos de artículos. No se realizaron técnicas de metanálisis, pero los resultados se compararon narrativamente. El protocolo de la RS se registró en PROSPERO (CRD42023426327). RESULTADOS: Se identificaron 917 artículos y se analizaron finalmente 20 que cumplían los criterios de inclusión. Los estudios incluidos contienen 33.182 HC procesados con 5.074 bacteriemias (15,3%). Once estudios fueron calificados de calidad alta, 7 moderada y 2 baja. El ABC-COR conseguida por el modelo de Shapiro varía de 0,71 a 0,83, con sensibilidad (Se) hasta del 98%, con especificidad (Es) (26% a 69%). Para los tres modelos que tienen validación interna y externa y una buena calidad metodológica, el modelo de Lee consigue un ABC-COR de 0,81 con Se: 68% y Es: 81%, el modelo 5MPB-Toledo consigue un ABC-COR entre 0,91 y 0,95, y el MPB-INFURG-SEMES obtiene una ABC-COR de 0,92 con una Se: 97% y Es: 76%. CONCLUSIONES: Los modelos 5MPB-Toledo y MPB-INFURG-SEMES representan herramientas útiles para la estratificación del riesgo real de bacteriemia en los pacientes atendidos en los SU.

Intubation Decision Based on Illness Severity and Mortality in COVID-19: An International Study.

OBJECTIVES: To evaluate the impact of intubation timing, guided by severity criteria, on mortality in critically ill COVID-19 patients, amidst existing uncertainties regarding optimal intubation practices. DESIGN: Prospective, multicenter, observational study conducted from February 1, 2020, to November 1, 2022. SETTING: Ten academic institutions in the United States and Europe. PATIENTS: Adults (≥ 18 yr old) confirmed with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and hospitalized specifically for COVID-19, requiring intubation postadmission. Exclusion criteria included patients hospitalized for non-COVID-19 reasons despite a positive SARS-CoV-2 test. INTERVENTIONS: Early invasive mechanical ventilation (EIMV) was defined as intubation in patients with less severe organ dysfunction (Sequential Organ Failure Assessment [SOFA] < 7 or Pao2/Fio2 ratio > 250), whereas late invasive mechanical ventilation (LIMV) was defined as intubation in patients with SOFA greater than or equal to 7 and Pao2/Fio2 ratio less than or equal to 250. MEASUREMENTS AND MAIN RESULTS: The primary outcome was mortality within 30 days of hospital admission. Among 4464 patients, 854 (19.1%) required mechanical ventilation (mean age 60 yr, 61.7% male, 19.3% Black). Of those, 621 (72.7%) were categorized in the EIMV group and 233 (27.3%) in the LIMV group. Death within 30 days after admission occurred in 278 patients (42.2%) in the EIMV and 88 patients (46.6%) in the LIMV group (p = 0.28). An inverse probability-of-treatment weighting analysis revealed a statistically significant association with mortality, with patients in the EIMV group being 32% less likely to die either within 30 days of admission (adjusted hazard ratio [HR] 0.68; 95% CI, 0.52-0.90; p = 0.008) or within 30 days after intubation irrespective of its timing from admission (adjusted HR 0.70; 95% CI, 0.51-0.90; p = 0.006). CONCLUSIONS: In severe COVID-19 cases, an early intubation strategy, guided by specific severity criteria, is associated with a reduced risk of death. These findings underscore the importance of timely intervention based on objective severity assessments.

Reliability of a point of care testing blood gas analyzer for measurement of lactate levels in cerebrospinal fluid.

Analysis of cerebrospinal fluid (CSF), including lactate, is key for diagnosis of acute meningitis. Since blood gas analyzers (BGA) enable rapid and safe blood-lactate measurements, we evaluated the reliability of RAPIDPoint 500 BGA to provide a fast and accurate measure of CSF lactate. In this study, CSF lactate levels were measured by a reference assay and on RAPIDPoint 500 BGA. Comparability was evaluated through difference analysis, using Bland Altman test, and linear regression analysis, using the Passing Bablok test. Agreement rate according to CSF lactate (≥3.5 and <3.5 mmol/L) was calculated using kappa (κ) statistic. Population study included 98 CSF samples. Concerning difference analysis, according to Bland-Altman test, bias was 0.13 mmol/L (CI 95%: -0.26 to 0.52 mmol/L. In regression analysis, according to Passing-Bablok equation a systematic difference between both assays was found. In concordance analysis, the interrate realibility was very high (κ: 0.964). According to our resuls, although a systematic difference was detected when lactate levels were measured on RAPIDPoint 500 BGA, the results from Bland-Altman test and the high agreement rate support that this POCT analyzer could be useful for a early and safe detection of patients with high probability of increased CSF lactate level.

Modelos predictivos de bacteriemia en el servicio de urgencias: revisión sistemática / Models to predict bacteremia in the emergency department: a systematic review

Objetivo. La obtención de hemocultivos (HC) se realiza en el 15% de los pacientes atendidos con sospecha de infección en los servicios de urgencias (SU) con una rentabilidad diagnóstica variable (2-20%). La mortalidad a 30 días de estos pacientes con bacteriemia es elevada, doble o triple que el resto con el mismo proceso. Así, encontrar un modelo predictivo de bacteriemia eficaz y aplicable en los SU sería muy importante. Clásicamente, el modelo de Shapiro ha sido la referencia en todo el mundo. El objetivo de esta revisión sistemática (RS) es comparar la capacidad para predecir bacteriemia en los SU de los distintos modelos predictivos publicados desde el año 2008 (fecha de publicación del modelo de Shapiro). Métodos. Se realiza una RS siguiendo la normativa PRISMA en las bases de datos de PubMed, Web of Science, EMBASE, Lilacs, Cochrane, Epistemonikos, Tripdatabase y ClinicalTrials.gov desde enero de 2008 hasta 31 mayo 2023 sin restricción de idiomas y utilizando una combinación de términos MESH: “Bacteremia/Bacteraemia/Blood Stream Infection”, “Prediction Model/Clinical Prediction Rule/Risk Prediction Model”, “Emergencies/Emergency/Emergency Department” y “Adults”. Se incluyeron estudios de cohortes observacionales (analíticos de rendimiento diagnóstico). Para valorar la calidad del método empleado y el riesgo de sesgos de los artículos incluidos se utilizó la NewcastleOttawa Scale (NOS). No se incluyeron estudios de casos y controles, revisiones narrativas y en otros tipos de artículos. No se realizaron técnicas de metanálisis, pero los resultados se compararon narrativamente. El protocolo de la RS se registró en PROSPERO (CRD42023426327). Resultados. Se identificaron 917 artículos y se analizaron finalmente 20 que cumplían los criterios de inclusión. Los estudios incluidos contienen 33.182 HC procesados con 5.074 bacteriemias (15,3%). Once estudios fueron calificados de calidad alta, 7 moderada y 2 baja... (AU)

Objective. Blood cultures are ordered in emergency departments for 15% of patients with suspected infection. The diagnostic yield varies from 2% to 20%. Thirty-day mortality in patients with bacteremia is high, doubling or tripling the rate in patients with the same infection but without bacteremia. Thus, finding an effective model to predict bacteremia that is applicable in emergency departments is an important goal. Shapiro’s model is the one traditionally used as a reference internationally. The aim of this systematic review was to compare the predictive power of bacteremia risk models published since 2008, when Shapiro’s model first appeared. Methods. We followed the recommendations of the Preferred Reporting Items for Systematic Reviews and MetaAnalyses (PRISMA) statement, searching in the following databases for articles published between January 2008 and May 31, 2023: PubMed, Web of Science, EMBASE, Lilacs, Cochrane, Epistemonikos, Trip Medical Database, and ClinicalTrials.gov. No language restrictions were specified. The search terms were the following Medical Subject Headings: bacteremia/bacteraemia/blood stream infection, prediction model/clinical prediction rule/risk prediction model, emergencies/emergency/emergency department, and adults. Observational cohort studies analyzing diagnostic yield were included; case-control studies, narrative reviews, and other types of articles were excluded. The Newcastle-Ottawa Scale was used to score quality and risk of bias in the included studies. The results were compared descriptively, without meta-analysis. The protocol was included in the PROSPERO register (CRD42023426327). Results. Twenty studies out of a total of 917 were found to meet the inclusion criteria. The included studies together analyzed 33 182 blood cultures, which detected 5074 cases of bacteremia (15.3%). Eleven studies were of high quality, 7 of moderate quality, and 2 of low quality... (AU)

Impact of New Onset Atrial Fibrillation in Hospitalized Patients with COVID-19. Results of the "ACO-VID" Registry.

OBJECTIVE: To assess the impact of new-onset atrial fibrillation (AF) on patients hospitalized with coronavirus disease 2019 (COVID-19). METHODS: Multicenter and retrospective study that included subjects >55 years hospitalized with COVID-19 from March to October 2020 in Spanish hospitals. Patients were divided into 3 groups (no AF, new-onset AF, and preexisting AF) and followed-up to 90 days. RESULTS: A total of 668 patients were included, of whom 162 (24.3%) had no AF, 107 (16.0%) new-onset AF and 399 (59.7%) preexisting AF. Compared to patients without AF, those patients with new-onset AF were older and had more comorbidities, but without differences with preexisting AF. During hospitalization, in the univariate analysis, compared to patients without AF, major bleeding and cardiovascular mortality were more frequent in patients with new-onset AF (10.3% vs 0.6%; P < .001; 2.8% vs 0.6%; P = .025, respectively), with a trend toward more stroke (1.9% vs 0%; P = .085). Outcomes were similar between AF groups, but the length of stay was greater in preexisting AF patients. Among patients with new-onset AF taking reduced doses of anticoagulant treatment was associated with higher risks of stroke and major bleeding. CONCLUSIONS: In COVID-19 hospitalized patients, new-onset AF may be associated with worse outcomes, but influenced by the dose of anticoagulants.

Discriminative performance of pancreatic stone protein in predicting ICU mortality and infection severity in adult patients with infection: a systematic review and individual patient level meta-analysis.

BACKGROUND: Several studies suggested pancreatic stone protein (PSP) as a promising biomarker to predict mortality among patients with severe infection. The objective of the study was to evaluate the performance of PSP in predicting intensive care unit (ICU) mortality and infection severity among critically ill adults admitted to the hospital for infection. METHODS: A systematic search across Cochrane Central Register of Controlled Trials and MEDLINE databases (1966 to February 2022) for studies on PSP published in English using 'pancreatic stone protein', 'PSP', 'regenerative protein', 'lithostatin' combined with 'infection' and 'sepsis' found 46 records. The search was restricted to the five trials that measured PSP using the enzyme-linked immunosorbent assay technique (ELISA). We used Bayesian hierarchical regression models for pooled estimates and to predict mortality or disease severity using PSP, C-Reactive Protein (CRP) and procalcitonin (PCT) as main predictor. We used statistical discriminative measures, such as the area under the receiver operating characteristic curve (AUC) and classification plots. RESULTS: Among the 678 patients included, the pooled ICU mortality was 17.8% (95% prediction interval 4.1% to 54.6%) with a between-study heterogeneity (I-squared 87%). PSP was strongly associated with ICU mortality (OR = 2.7, 95% credible interval (CrI) [1.3-6.0] per one standard deviation increase; age, gender and sepsis severity adjusted OR = 1.5, 95% CrI [0.98-2.8]). The AUC was 0.69 for PSP 95% confidence interval (CI) [0.64-0.74], 0.61 [0.56-0.66] for PCT and 0.52 [0.47-0.57] for CRP. The sensitivity was 0.96, 0.52, 0.30 for risk thresholds 0.1, 0.2 and 0.3; respective false positive rate values were 0.84, 0.25, 0.10. CONCLUSIONS: We found that PSP showed a very good discriminative ability for both investigated study endpoints ICU mortality and infection severity; better in comparison to CRP, similar to PCT. Combinations of biomarkers did not improve their predictive ability.

Effects of Etomidate on Postintubation Hypotension, Inflammatory Markers, and Mortality in Critically Ill Patients with COVID-19: An International, Multicenter, Retrospective Study.

OBJECTIVE: To evaluate the association of etomidate with postintubation hypotension, inflammation, and mortality in critically ill patients with COVID-19. DESIGN: International, multicenter, retrospective study. PARTICIPANTS: Critically ill patients hospitalized specifically for COVID-19 from three major academic institutions in the US and Europe. MAIN OUTCOME AND MEASURES: Patients were allocated into the etomidate (ET) group or another induction agent (OA) group. The primary outcome was postintubation hypotension. Secondary outcomes included postintubation inflammatory status, in-hospital mortality, and mortality at 30 days. RESULTS: 171 patients with a median age of 68 (IQR 58-73) years were included (ET, n = 98; OA, n = 73). Etomidate was associated with lower postintubation mean arterial pressure [74.33 (64-85) mm Hg versus 81.84 (69.75-94.25) mm Hg, p = 0.005] compared to other agents. No statistically significant differences were generally observed in inflammatory markers between the two groups at 7- and 14-days after admission to the intensive care unit. In-hospital mortality [77 (79%) versus 41 (56%), p = 0.003] and mortality at 30-days [78 (80%) versus 43 (59%), p = 0.006] were higher in the ET group. In multivariate logistic regression analysis, only etomidate (p = 0.009) and postintubation mean arterial pressure (p < 0.001) had a statistically significant effect on mortality, in contrast to stress-dose steroids (p = 0.301), after adjusting for creatinine (p = 0.695), blood urea nitrogen (p = 0.153), age (p = 0.055), oxygen saturation of hemoglobin (SpO2) (p = 0.941), and fraction of inspired oxygen (FiO2) (p = 0.712). CONCLUSIONS: Administration of a single-bolus dose of etomidate in critically ill patients with COVID-19 is associated with lower postintubation mean arterial pressure and higher in-hospital and 30-day mortality compared to other induction agents.

Identification of COVID-19 patients at risk of hospital admission and mortality: a European multicentre retrospective analysis of mid-regional pro-adrenomedullin.

BACKGROUND: Mid-Regional pro-Adrenomedullin (MR-proADM) is an inflammatory biomarker that improves the prognostic assessment of patients with sepsis, septic shock and organ failure. Previous studies of MR-proADM have primarily focussed on bacterial infections. A limited number of small and monocentric studies have examined MR-proADM as a prognostic factor in patients infected with SARS-CoV-2, however there is need for multicenter validation. An evaluation of its utility in predicting need for hospitalisation in viral infections was also performed. METHODS: An observational retrospective analysis of 1861 patients, with SARS-CoV-2 confirmed by RT-qPCR, from 10 hospitals across Europe was performed. Biomarkers, taken upon presentation to Emergency Departments (ED), clinical scores, patient demographics and outcomes were collected. Multiclass random forest classifier models were generated as well as calculation of area under the curve analysis. The primary endpoint was hospital admission with and without death. RESULTS: Patients suitable for safe discharge from Emergency Departments could be identified through an MR-proADM value of ≤ 1.02 nmol/L in combination with a CRP (C-Reactive Protein) of ≤ 20.2 mg/L and age ≤ 64, or in combination with a SOFA (Sequential Organ Failure Assessment) score < 2 if MR-proADM was ≤ 0.83 nmol/L regardless of age. Those at an increased risk of mortality could be identified upon presentation to secondary care with an MR-proADM value of > 0.85 nmol/L, in combination with a SOFA score ≥ 2 and LDH > 720 U/L, or in combination with a CRP > 29.26 mg/L and age ≤ 64, when MR-proADM was > 1.02 nmol/L. CONCLUSIONS: This international study suggests that for patients presenting to the ED with confirmed SARS-CoV-2 infection, MR-proADM in combination with age and CRP or with the patient's SOFA score could identify patients at low risk where outpatient treatment may be safe.

Circulating suPAR associates with severity and in-hospital progression of COVID-19.

BACKGROUND: COVID-19 disease progression is characterized by hyperinflammation and risk stratification may aid in early aggressive treatment and advanced planning. The aim of this study was to assess whether suPAR and other markers measured at hospital admission can predict the severity of COVID-19. METHODS: The primary outcome measure in this international, multi-centre, prospective, observational study with adult patients hospitalized primarily for COVID-19 was the association of WHO Clinical Progression Scale (WHO-CPS) with suPAR, ferritin, CRP, albumin, LDH, eGFR, age, procalcitonin, and interleukin-6. Admission plasma suPAR levels were determined using the suPARnostic® ELISA and suPARnostic® Turbilatex assays. RESULTS: Seven hundred and sixty-seven patients, 440 (57.4%) males and 327 (42.6%) females, were included with a median age of 64 years. Log-suPAR levels significantly correlated with WHO-CPS score, with each doubling of suPAR increasing the score by one point (p < .001). All the other markers were also correlated with WHO-CPS score. Admission suPAR levels were significantly lower in survivors (7.10 vs. 9.63, 95% CI 1.47-3.59, p < .001). A linear model (SALGA) including suPAR, serum albumin, serum lactate dehydrogenase, eGFR, and age can best estimate the WHO-CPS score and survival. Combining all five parameters in the SALGA model can improve the accuracy of discrimination with an AUC of 0.80 (95% CI: 0.759-0.836). CONCLUSIONS: suPAR levels significantly correlated with WHO-CPS score, with each doubling of suPAR increasing the score by one point. The SALGA model may serve as a quick tool for predicting disease severity and survival at admission.

Dabigatran in patients with atrial fibrillation after COVID-19 hospitalization: an update of the ANIBAL protocol.

COVID-19 increases the risk of atrial fibrillation (AF) and thrombotic complications, particularly in severe cases, leading to higher mortality rates. Anticoagulation is the cornerstone to reduce thromboembolic risk in patients with AF. Considering the risk of hepatotoxicity in patients with severe COVID-19 as well as the risk of drug-drug interactions, drug-induced hepatotoxicity and bleeding, the ANIBAL protocol was developed to facilitate the anticoagulation approach at discharge after COVID-19 hospitalization. However, since the publication of the original algorithm, relevant changes have occurred. First, treatment of COVID-19 pneumonia has been modified with the use of dexamethasone or remdesivir during the first week in patients that require oxygen therapy, and of dexamethasone and/or tocilizumab or baricitinib during the second week in patients that necessitate supplementary oxygen or with a high inflammation state, respectively. On the other hand, metabolic syndrome is common in patients with AF as well as metabolic-associated fatty liver disease, and this could negatively impact the prognosis of patients with COVID-19, including high transaminase levels in patients treated with immunomodulators. The EHRA guidelines update also introduce some interesting changes in drug-drug interaction patterns with the reduction of the level of the interaction with dexamethasone, which is of paramount importance in this clinical context. Considering the new information, the protocol, named ANIBAL II, has been updated. In this new protocol, the anticoagulant of choice in patients with AF after COVID-19 hospitalization is provided according to three scenarios: with/without dexamethasone treatment at discharge and normal hepatic function, transaminases ≤2 times the upper limit of normal, or transaminases >2 times the upper limit of normal.

Circulating levels of calprotectin, a signature of neutrophil activation in prediction of severe respiratory failure in COVID-19 patients: a multicenter, prospective study (CalCov study).

OBJECTIVE: Severe COVID-19 is characterized by a dysregulated immune response in which neutrophils play a critical role. Calprotectin reflects neutrophil activation and is involved in the self-amplifying thrombo-inflammatory storm in severe COVID-19. We aimed to evaluate the role of calprotectin in early prediction of severity in COVID-19 patients. METHODS: This was a multicenter prospective observational study enrolling consecutive adult COVID-19 patients. On arrival to emergency department, blood samples were collected for laboratory tests, including serum calprotectin. The primary outcome was severe respiratory failure requiring invasive mechanical ventilation and the secondary outcome was need for Intensive Care Unit (ICU) admission. RESULTS: Study population included 395 patients, 57 (14.4%) required invasive mechanical ventilation and 100 (25.3%) were admitted to ICU. Median serum calprotectin levels were significantly higher in intubated (3.73 mg/L vs. 2.63 mg/L; p < 0.001) and ICU patients (3.48 mg/L vs. 2.60 mg/L; p = 0.001). Calprotectin showed a significant accuracy to predict the need for invasive mechanical ventilation (ROC AUC 0.723) and ICU admission (ROC AUC 0.650). In multivariate analysis, serum calprotectin was an independent predictor of invasive mechanical ventilation (OR 1.161) and ICU admission (OR 1.068). CONCLUSION: Serum calprotectin can be used as an early predictor of severity in COVID-19 patients.

Role of Pancreatic Stone Protein as an Early Biomarker for Risk Stratification of Acute Pancreatitis.

BACKGROUND: Early risk stratification of acute pancreatitis is crucial to improve clinical outcomes. The objective of this study was to evaluate the ability of pancreatic stone protein (PSP) to predict acute pancreatitis severity and to compare it with the biomarkers and severity scores currently used for that purpose. PATIENTS AND METHODS: Prospective single-center observational study enrolling 268 adult patients with acute pancreatitis. Biomarkers including PSP were measured upon admission to the Emergency Department and severity scores as SOFA, PANC-3, and BISAP were computed. Patients were classified into mild-moderate (non-severe) and severe acute pancreatitis according to the Determinant-Based Classification Criteria. Area under the curve (AUC) and regression analysis were used to analyze the discrimination abilities and the association of biomarkers and scores with severity. RESULTS: Two hundred and thirty-five patients (87.7%) were classified as non-severe and 33 (12.3%) as severe acute pancreatitis. Median [IQR] PSP was increased in patients with severe acute pancreatitis (890 µg/L [559-1142] vs. 279 µg/L [141-496]; p < 0.001) and it was the best predictor (ROC AUC: 0.827). In multivariate analysis, PSP and urea were the only independent predictors for severe acute pancreatitis and a model combining them both ("biomarker model") showed an AUC of 0.841 for prediction of severe acute pancreatitis, higher than the other severity scores. CONCLUSIONS: PSP is a promising biomarker for predicting the severity of acute pancreatitis upon admission. A model combining PSP and urea might further constitute a potential tool for early risk stratification of this disease.

Harmonized D-dimer levels upon admission for prognosis of COVID-19 severity: Results from a Spanish multicenter registry (BIOCOVID-Spain study).

Coagulopathy is a key feature of COVID-19 and D-dimer has been reported as a predictor of severity. However, because D-dimer test results vary considerably among assays, resolving harmonization issues is fundamental to translate findings into clinical practice. In this retrospective multicenter study (BIOCOVID study), we aimed to analyze the value of harmonized D-dimer levels upon admission for the prediction of in-hospital mortality in COVID-19 patients. All-cause in-hospital mortality was defined as endpoint. For harmonization of D-dimer levels, we designed a model based on the transformation of method-specific regression lines to a reference regression line. The ability of D-dimer for prediction of death was explored by receiver operating characteristic curves analysis and the association with the endpoint by Cox regression analysis. Study population included 2663 patients. In-hospital mortality rate was 14.3%. Harmonized D-dimer upon admission yielded an area under the curve of 0.66, with an optimal cut-off value of 0.945 mg/L FEU. Patients with harmonized D-dimer ≥ 0.945 mg/L FEU had a higher mortality rate (22.4% vs. 9.2%; p < 0.001). D-dimer was an independent predictor of in-hospital mortality, with an adjusted hazard ratio of 1.709. This is the first study in which a harmonization approach was performed to assure comparability of D-dimer levels measured by different assays. Elevated D-dimer levels upon admission were associated with a greater risk of in-hospital mortality among COVID-19 patients, but had limited performance as prognostic test.

Circulating MR-proADM levels, as an indicator of endothelial dysfunction, for early risk stratification of mid-term mortality in COVID-19 patients.

OBJECTIVES: Thromboinflammation, resulting from a complex interaction between thrombocytopathy, coagulopathy, and endotheliopathy, contributes to increased mortality in COVID-19 patients. MR-proADM, as a surrogate of adrenomedullin system disruption, leading to endothelial damage, has been reported as a promising biomarker for short-term prognosis. We evaluated the role of MR-proADM in the mid-term mortality in COVID-19 patients. METHODS: A prospective, observational study enrolling COVID-19 patients from August to October 2020. A blood sample for laboratory test analysis was drawn on arrival in the emergency department. The primary endpoint was 90-day mortality. The area under the curve (AUC) and Cox regression analyses were used to assess discriminatory ability and association with the endpoint. RESULTS: A total of 359 patients were enrolled, and the 90-day mortality rate was 8.9%. ROC AUC for MR-proADM predicting 90-day mortality was 0.832. An optimal cutoff of 0.80 nmol/L showed a sensitivity of 96.9% and a specificity of 58.4%, with a negative predictive value of 99.5%. Circulating MR-proADM levels (inverse transformed), after adjusting by a propensity score including eleven potential confounders, were an independent predictor of 90-day mortality (HR: 0.162 [95% CI: 0.043-0.480]) CONCLUSIONS: Our data confirm that MR-proADM has a role in the mid-term prognosis of COVID-19 patients and might assist physicians with risk stratification.

The importance of being external. methodological insights for the external validation of machine learning models in medicine.

Background and Objective Medical machine learning (ML) models tend to perform better on data from the same cohort than on new data, often due to overfitting, or co-variate shifts. For these reasons, external validation (EV) is a necessary practice in the evaluation of medical ML. However, there is still a gap in the literature on how to interpret EV results and hence assess the robustness of ML models. METHODS: We fill this gap by proposing a meta-validation method, to assess the soundness of EV procedures. In doing so, we complement the usual way to assess EV by considering both dataset cardinality, and the similarity of the EV dataset with respect to the training set. We then investigate how the notions of cardinality and similarity can be used to inform on the reliability of a validation procedure, by integrating them into two summative data visualizations. RESULTS: We illustrate our methodology by applying it to the validation of a state-of-the-art COVID-19 diagnostic model on 8 EV sets, collected across 3 different continents. The model performance was moderately impacted by data similarity (Pearson ρ = 0.38, p< 0.001). In the EV, the validated model reported good AUC (average: 0.84), acceptable calibration (average: 0.17) and utility (average: 0.50). The validation datasets were adequate in terms of dataset cardinality and similarity, thus suggesting the soundness of the results. We also provide a qualitative guideline to evaluate the reliability of validation procedures, and we discuss the importance of proper external validation in light of the obtained results. CONCLUSIONS: In this paper, we propose a novel, lean methodology to: 1) study how the similarity between training and validation sets impacts the generalizability of a ML model; 2) assess the soundness of EV evaluations along three complementary performance dimensions: discrimination, utility and calibration; 3) draw conclusions on the robustness of the model under validation. We applied this methodology to a state-of-the-art model for the diagnosis of COVID-19 from routine blood tests, and showed how to interpret the results in light of the presented framework.