Impact of macrolide treatment on long-term mortality in patients admitted to the ICU due to CAP: a targeted maximum likelihood estimation and survival analysis.

INTRODUCTION: Patients with community-acquired pneumonia (CAP) admitted to the intensive care unit (ICU) have high mortality rates during the acute infection and up to ten years thereafter. Recommendations from international CAP guidelines include macrolide-based treatment. However, there is no data on the long-term outcomes of this recommendation. Therefore, we aimed to determine the impact of macrolide-based therapy on long-term mortality in this population. METHODS: Registered patients in the MIMIC-IV database 16 years or older and admitted to the ICU due to CAP were included. Multivariate analysis, targeted maximum likelihood estimation (TMLE) to simulate a randomised controlled trial, and survival analyses were conducted to test the effect of macrolide-based treatment on mortality six-month (6 m) and twelve-month (12 m) after hospital admission. A sensitivity analysis was performed excluding patients with Pseudomonas aeruginosa or MRSA pneumonia to control for Healthcare-Associated Pneumonia (HCAP). RESULTS: 3775 patients were included, and 1154 were treated with a macrolide-based treatment. The non-macrolide-based group had worse long-term clinical outcomes, represented by 6 m [31.5 (363/1154) vs 39.5 (1035/2621), p < 0.001] and 12 m mortality [39.0 (450/1154) vs 45.7 (1198/2621), p < 0.001]. The main risk factors associated with long-term mortality were Charlson comorbidity index, SAPS II, septic shock, and respiratory failure. Macrolide-based treatment reduced the risk of dying at 6 m [HR (95% CI) 0.69 (0.60, 0.78), p < 0.001] and 12 m [0.72 (0.64, 0.81), p < 0.001]. After TMLE, the protective effect continued with an additive effect estimate of - 0.069. CONCLUSION: Macrolide-based treatment reduced the hazard risk of long-term mortality by almost one-third. This effect remains after simulating an RCT with TMLE and the sensitivity analysis for the HCAP classification.

Effects and safety of hypertonic saline combined with airway clearance in non-hospitalized children with recurrent wheezing.

Background: The International Study of Wheezing in Infants defines recurrent wheezing as the presence of three or more medically documented episodes of wheezing within one year. To date, there is no evidence on the use of hypertonic saline (HS) combined with airway clearance techniques (ACT) for children with recurrent wheezing treated in an outpatient setting. Therefore, this is the first study to explore the use of such interventions in infants with recurrent wheezing. Objectives: To evaluate the effects and safety of a three-month protocol including HS and ACT for non-hospitalized infants with recurrent wheezing. Methods: Randomized, double-blind, controlled trial, including outpatient infants with recurrent wheezing. Children were randomized to either 3% HS or 0.9% saline groups and were treated with bronchodilator and nebulized with the respective solutions before ACT. The primary outcome was the Wang score. Secondary outcomes included the number of hospitalizations and respiratory crisis, need for rescue medication, and school absences. All variables were measured during the three previous months from inclusion and during intervention period. The study protocol was registered at ClinicalTrials.gov (NCT04331496) on March, 31, 2020. Results: Forty children were included. Regarding immediate effects, significant differences (p<0.001) were found for time, but not for group or interaction (group × time), in all outcome variables (increase in SpO2, decrease in heart and respiratory rate, wheezing episodes, retraction, and Wang score). Comparing the previous three months with the study period, there were significant differences in both groups for the severity of crisis (p<0.001) and medication steps (p=0.002). Conclusion: A three-month protocol including HS and ACT for outpatient infants with recurrent wheezing was safe and reduced morbidity. No differences were found between the use of HS and 0.9% saline.

Deploying unsupervised clustering analysis to derive clinical phenotypes and risk factors associated with mortality risk in 2022 critically ill patients with COVID-19 in Spain.

BACKGROUND: The identification of factors associated with Intensive Care Unit (ICU) mortality and derived clinical phenotypes in COVID-19 patients could help for a more tailored approach to clinical decision-making that improves prognostic outcomes. METHODS: Prospective, multicenter, observational study of critically ill patients with confirmed COVID-19 disease and acute respiratory failure admitted from 63 ICUs in Spain. The objective was to utilize an unsupervised clustering analysis to derive clinical COVID-19 phenotypes and to analyze patient's factors associated with mortality risk. Patient features including demographics and clinical data at ICU admission were analyzed. Generalized linear models were used to determine ICU morality risk factors. The prognostic models were validated and their performance was measured using accuracy test, sensitivity, specificity and ROC curves. RESULTS: The database included a total of 2022 patients (mean age 64 [IQR 5-71] years, 1423 (70.4%) male, median APACHE II score (13 [IQR 10-17]) and SOFA score (5 [IQR 3-7]) points. The ICU mortality rate was 32.6%. Of the 3 derived phenotypes, the A (mild) phenotype (537; 26.7%) included older age (< 65 years), fewer abnormal laboratory values and less development of complications, B (moderate) phenotype (623, 30.8%) had similar characteristics of A phenotype but were more likely to present shock. The C (severe) phenotype was the most common (857; 42.5%) and was characterized by the interplay of older age (> 65 years), high severity of illness and a higher likelihood of development shock. Crude ICU mortality was 20.3%, 25% and 45.4% for A, B and C phenotype respectively. The ICU mortality risk factors and model performance differed between whole population and phenotype classifications. CONCLUSION: The presented machine learning model identified three clinical phenotypes that significantly correlated with host-response patterns and ICU mortality. Different risk factors across the whole population and clinical phenotypes were observed which may limit the application of a "one-size-fits-all" model in practice.

Methods and measures to quantify ICU patient heterogeneity.

Patients in intensive care units are heterogeneous and the daily prediction of their days to discharge (DTD) a complex task that practitioners and computers are not always able to solve satisfactorily. In order to make more precise DTD predictors, it is necessary to have tools for the analysis of the heterogeneity of the patients. Unfortunately, the number of publications in this field is almost non-existent. In order to alleviate this lack of tools, we propose four methods and their corresponding measures to quantify the heterogeneity of intensive patients in the process of determining the DTD. These new methods and measures have been tested with patients admitted over four years to a tertiary hospital in Spain. The results deepen the understanding of the intensive patient and can serve as a basis for the construction of better DTD predictors.

Factors Leading to the Loss of Natural Elite Control of HIV-1 Infection.

HIV-1 elite controllers (EC) maintain undetectable viral loads (VL) in the absence of antiretroviral treatment. However, these subjects have heterogeneous clinical outcomes, including a proportion that loses HIV-1 control over time. In this work, we compared, in a longitudinal design, transient EC, analyzed before and after the loss of virological control, with persistent EC. The aim was to identify factors leading to the loss of natural virological control of HIV-1 infection with a longitudinal retrospective study design. Gag-specific T-cell responses were assessed by in vitro intracellular polycytokine production quantified by flow cytometry. Viral diversity determinations and sequence dating were performed in proviral DNA by PCR amplification at limiting dilution of env and gag genes. The expression profile of 70 serum cytokines and chemokines was assessed by multiplex immunoassays. We identified transient EC as subjects with low Gag-specific T-cell polyfunctionality, high viral diversity, and high proinflammatory cytokine levels before the loss of control. Gag-specific T-cell polyfunctionality was inversely associated with viral diversity in transient controllers before the loss of control (r = -0.8; P = 0.02). RANTES was a potential biomarker of transient control. This study identified virological and immunological factors, including inflammatory biomarkers associated with two different phenotypes within EC. These results may allow a more accurate definition of EC, which could help in better clinical management of these individuals and in the development of future curative approaches.IMPORTANCE There is a rare group of HIV-infected patients who have the extraordinary capacity to maintain undetectable viral load levels in the absence of antiretroviral treatment, the so-called HIV-1 elite controllers (EC). However, there is a proportion within these subjects that eventually loses this capability. In this work, we found differences in virological and immune factors, including soluble inflammatory biomarkers, between subjects with persistent control of viral replication and EC that will lose virological control. The identification of these factors could be a key point for a right medical care of those EC who are going to lose natural control of viral replication and for the design of future immunotherapeutic strategies using as a model the natural persistent control of HIV infection.

rDolphin: a GUI R package for proficient automatic profiling of 1D 1H-NMR spectra of study datasets.

INTRODUCTION: Adoption of automatic profiling tools for 1H-NMR-based metabolomic studies still lags behind other approaches in the absence of the flexibility and interactivity necessary to adapt to the properties of study data sets of complex matrices. OBJECTIVES: To provide an open source tool that fully integrates these needs and enables the reproducibility of the profiling process. METHODS: rDolphin incorporates novel techniques to optimize exploratory analysis, metabolite identification, and validation of profiling output quality. RESULTS: The information and quality achieved in two public datasets of complex matrices are maximized. CONCLUSION: rDolphin is an open-source R package ( http://github.com/danielcanueto/rDolphin ) able to provide the best balance between accuracy, reproducibility and ease of use.

The Experience of Implementing Urban HEART Barcelona: a Tool for Action.

Urban Health Equity Assessment and Response Tool (HEART) is a tool developed by the World Health Organization whose objective is to provide evidence on urban health inequalities so as to help to decide the best interventions aimed to promote urban health equity. The aim of this paper is to describe the experience of implementing Urban HEART in Barcelona city, both the adaptation of Urban HEART to the city of Barcelona, its use as a means of identifying and monitoring health inequalities among city neighbourhoods, and the difficulties and barriers encountered throughout the process. Although ASPB public health technicians participated in the Urban HEART Advisory Group, had large experience in health inequalities analysis and research and showed interest in implementing the tool, it was not until 2015, when the city council was governed by a new left-wing party for which reducing health inequalities was a priority that Urban HEART could be used. A provisional matrix was developed, including both health and health determinant indicators, which allowed to show how some neighbourhoods in the city systematically fare worse for most of the indicators while others systematically fare better. It also allowed to identify 18 neighbourhoods-those which fared worse in most indicators-which were considered a priority for intervention, which entered the Health in the Barcelona Neighbourhoods programme and the Neighbourhoods Plan. This provisional version was reviewed and improved by the Urban HEART Barcelona Working Group. Technicians with experience in public health and/or in indicator and database management were asked to indicate suitability and relevance from a list of potential indicators. The definitive Urban HEART Barcelona version included 15 indicators from the five Urban HEART domains and improved the previous version in several requirements. Several barriers were encountered, such as having to estimate indicators in scarcely populated areas or finding adequate indicators for the physical context domain. In conclusion, the Urban HEART tool allowed to identify urban inequalities in the city of Barcelona and to include health inequalities in the public debate. It also allowed to reinforce the community health programme Health in the Barcelona Neighbourhoods as well as other city programmes aimed at reducing health inequalities. A strong political will is essential to place health inequalities in the political agenda and implement policies to tackle them.

Positional Enrichment by Proton Analysis (PEPA): A One-Dimensional 1 H-NMR Approach for 13 C Stable Isotope Tracer Studies in Metabolomics.

A novel metabolomics approach for NMR-based stable isotope tracer studies called PEPA is presented, and its performance validated using human cancer cells. PEPA detects the position of carbon label in isotopically enriched metabolites and quantifies fractional enrichment by indirect determination of 13 C-satellite peaks using 1D-1 H-NMR spectra. In comparison with 13 C-NMR, TOCSY and HSQC, PEPA improves sensitivity, accelerates the elucidation of 13 C positions in labeled metabolites and the quantification of the percentage of stable isotope enrichment. Altogether, PEPA provides a novel framework for extending the high-throughput of 1 H-NMR metabolic profiling to stable isotope tracing in metabolomics, facilitating and complementing the information derived from 2D-NMR experiments and expanding the range of isotopically enriched metabolites detected in cellular extracts.

Dolphin: a tool for automatic targeted metabolite profiling using 1D and 2D (1)H-NMR data.

One of the main challenges in nuclear magnetic resonance (NMR) metabolomics is to obtain valuable metabolic information from large datasets of raw NMR spectra in a high throughput, automatic, and reproducible way. To date, established software packages used to match and quantify metabolites in NMR spectra remain mostly manually operated, leading to low resolution results and subject to inconsistencies not attributable to the NMR technique itself. Here, we introduce a new software package, called Dolphin, able to automatically quantify a set of target metabolites in multiple sample measurements using an approach based on 1D and 2D NMR techniques to overcome the inherent limitations of 1D (1)H-NMR spectra in metabolomics. Dolphin takes advantage of the 2D J-resolved NMR spectroscopy signal dispersion to avoid inconsistencies in signal position detection, enhancing the reliability and confidence in metabolite matching. Furthermore, in order to improve accuracy in quantification, Dolphin uses 2D NMR spectra to obtain additional information on all neighboring signals surrounding the target metabolite. We have compared the targeted profiling results of Dolphin, recorded from standard biological mixtures, with those of two well established approaches in NMR metabolomics. Overall, Dolphin produced more accurate results with the added advantage of being a fully automated and high throughput processing package.

Assessing the impact of real-time random safety audits through full propensity score matching on reliable data from the clinical information system.

BACKGROUND: Evidence-based care processes are not always applied at the bedside in critically ill patients. Numerous studies have assessed the impact of checklists and related strategies on the process of care and patient outcomes. We aimed to evaluate the effects of real-time random safety audits on process-of-care and outcome variables in critical care patients. METHODS: This prospective study used data from the clinical information system to evaluate the impact of real-time random safety audits targeting 32 safety measures in two intensive care units during a 9-month period. We compared endpoints between patients attended with safety audits and those not attended with safety audits. The primary endpoint was mortality, measured by Cox hazard regression after full propensity-score matching. Secondary endpoints were the impact on adherence to process-of-care measures and on quality indicators. RESULTS: We included 871 patients; 228 of these were attended in ≥ 1 real-time random safety audits. Safety audits were carried out on 390 patient-days; most improvements in the process of care were observed in safety measures related to mechanical ventilation, renal function and therapies, nutrition, and clinical information system. Although the group of patients attended in safety audits had more severe disease at ICU admission [APACHE II score 21 (16-27) vs. 20 (15-25), p = 0.023]; included a higher proportion of surgical patients [37.3 % vs. 26.4 %, p = 0.003] and a higher proportion of mechanically ventilated patients [72.8 % vs. 40.3 %, p < 0.001]; averaged more days on mechanical ventilation, central venous catheter, and urinary catheter; and had a longer ICU stay [12.5 (5.5-23.3) vs. 2.9 (1.7-5.9), p < 0.001], ICU mortality did not differ significantly between groups (19.3 % vs. 18.8 % in the group without safety rounds). After full propensity-score matching, Cox hazard regression analysis showed real-time random safety audits were associated with a lower risk of mortality throughout the ICU stay (HR 0.31; 95 %CI 0.20-0.47). CONCLUSIONS: Real-time random safety audits are associated with a reduction in the risk of ICU mortality. Exploiting data from the clinical information system is useful in assessing the impact of them on the care process, quality indicators, and mortality.

Association of obesity on the outcome of critically ill patients affected by COVID-19.

OBJECTIVE: To evaluate the impact of obesity on ICU mortality. DESIGN: Observational, retrospective, multicentre study. SETTING: Intensive Care Unit (ICU). PATIENTS: Adults patients admitted with COVID-19 and respiratory failure. INTERVENTIONS: None. PRIMARY VARIABLES OF INTEREST: Collected data included demographic and clinical characteristics, comorbidities, laboratory tests and ICU outcomes. Body mass index (BMI) impact on ICU mortality was studied as (1) a continuous variable, (2) a categorical variable obesity/non-obesity, and (3) as categories defined a priori: underweight, normal, overweight, obesity and Class III obesity. The impact of obesity on mortality was assessed by multiple logistic regression and Smooth Restricted cubic (SRC) splines for Cox hazard regression. RESULTS: 5,206 patients were included, 20 patients (0.4%) as underweight, 887(17.0%) as normal, 2390(46%) as overweight, 1672(32.1) as obese and 237(4.5%) as class III obesity. The obesity group patients (n = 1909) were younger (61 vs. 65 years, p < 0.001) and with lower severity scores APACHE II (13 [9-17] vs. 13[10-17, p < 0.01) than non-obese. Overall ICU mortality was 28.5% and not different for obese (28.9%) or non-obese (28.3%, p = 0.65). Only Class III obesity (OR = 2.19, 95%CI 1.44-3.34) was associated with ICU mortality in the multivariate and SRC analysis. CONCLUSIONS: COVID-19 patients with a BMI > 40 are at high risk of poor outcomes in the ICU. An effective vaccination schedule and prolonged social distancing should be recommended.

Applicability of an unsupervised cluster model developed on first wave COVID-19 patients in second/third wave critically ill patients.

OBJECTIVE: To validate the unsupervised cluster model (USCM) developed during the first pandemic wave in a cohort of critically ill patients from the second and third pandemic waves. DESIGN: Observational, retrospective, multicentre study. SETTING: Intensive Care Unit (ICU). PATIENTS: Adult patients admitted with COVID-19 and respiratory failure during the second and third pandemic waves. INTERVENTIONS: None. MAIN VARIABLES OF INTEREST: Collected data included demographic and clinical characteristics, comorbidities, laboratory tests and ICU outcomes. To validate our original USCM, we assigned a phenotype to each patient of the validation cohort. The performance of the classification was determined by Silhouette coefficient (SC) and general linear modelling. In a post-hoc analysis we developed and validated a USCM specific to the validation set. The model's performance was measured using accuracy test and area under curve (AUC) ROC. RESULTS: A total of 2330 patients (mean age 63 [53-82] years, 1643 (70.5%) male, median APACHE II score (12 [9-16]) and SOFA score (4 [3-6]) were included. The ICU mortality was 27.2%. The USCM classified patients into 3 clinical phenotypes: A (n = 1206 patients, 51.8%); B (n = 618 patients, 26.5%), and C (n = 506 patients, 21.7%). The characteristics of patients within each phenotype were significantly different from the original population. The SC was -0.007 and the inclusion of phenotype classification in a regression model did not improve the model performance (0.79 and 0.78 ROC for original and validation model). The post-hoc model performed better than the validation model (SC -0.08). CONCLUSION: Models developed using machine learning techniques during the first pandemic wave cannot be applied with adequate performance to patients admitted in subsequent waves without prior validation.

Negative predictive value of procalcitonin to rule out bacterial respiratory co-infection in critical covid-19 patients.

BACKGROUND: Procalcitonin (PCT) and C-Reactive Protein (CRP) are useful biomarkers to differentiate bacterial from viral or fungal infections, although the association between them and co-infection or mortality in COVID-19 remains unclear. METHODS: The study represents a retrospective cohort study of patients admitted for COVID-19 pneumonia to 84 ICUs from ten countries between (March 2020-January 2021). Primary outcome was to determine whether PCT or CRP at admission could predict community-acquired bacterial respiratory co-infection (BC) and its added clinical value by determining the best discriminating cut-off values. Secondary outcome was to investigate its association with mortality. To evaluate the main outcome, a binary logistic regression was performed. The area under the curve evaluated diagnostic performance for BC prediction. RESULTS: 4635 patients were included, 7.6% fulfilled BC diagnosis. PCT (0.25[IQR 0.1-0.7] versus 0.20[IQR 0.1-0.5]ng/mL, p<0.001) and CRP (14.8[IQR 8.2-23.8] versus 13.3 [7-21.7]mg/dL, p=0.01) were higher in BC group. Neither PCT nor CRP were independently associated with BC and both had a poor ability to predict BC (AUC for PCT 0.56, for CRP 0.54). Baseline values of PCT<0.3ng/mL, could be helpful to rule out BC (negative predictive value 91.1%) and PCT≥0.50ng/mL was associated with ICU mortality (OR 1.5,p<0.001). CONCLUSIONS: These biomarkers at ICU admission led to a poor ability to predict BC among patients with COVID-19 pneumonia. Baseline values of PCT<0.3ng/mL may be useful to rule out BC, providing clinicians a valuable tool to guide antibiotic stewardship and allowing the unjustified overuse of antibiotics observed during the pandemic, additionally PCT≥0.50ng/mL might predict worsening outcomes.

Comparison of in vivo acute stent recoil between the bioresorbable everolimus-eluting coronary scaffolds (revision 1.0 and 1.1) and the metallic everolimus-eluting stent.

OBJECTIVES: The ABSORB cohort A trial using the bioresorbable everolimus-eluting scaffold (BVS revision 1.0, Abbott Vascular) demonstrated a slightly higher acute recoil with BVS than with metallic stents. To reinforce the mechanical strength of the scaffold, the new BVS scaffold (revision 1.1) with modified strut design was developed and tested in the ABSORB cohort B trial. This study sought to evaluate and compare the in vivo acute scaffold recoil of the BVS revision 1.0 in ABSORB cohort A and the BVS revision 1.1 in ABSORB cohort B with the historical recoil of the XIENCE V® everolimus-eluting metal stent (EES, SPIRIT I and II). METHODS: In the ABSORB cohort B trial, 101 patients with one or two de-novo lesions were enrolled at 10 sites. In ABSORB cohort A, 27 patients treated with a BVS 1.0 were analyzed and compared with EES. Acute absolute recoil, assessed by quantitative coronary angiography, was defined as the difference between mean diameter of the last inflated balloon at the highest pressure (X) and mean lumen diameter of the stent immediately after the last balloon deflation (Y). Acute percent recoil was defined as (X - Y)/X and expressed as a percentage. RESULTS: Out of 101 patients enrolled in the ABSORB cohort B trial, 88 patients are available for complete analysis of acute recoil. Absolute recoil of BVS 1.1 (0.19 ± 0.18 mm) was numerically higher than metallic EES (vs. 0.13 ± 0.21 mm) and similar to BVS 1.0 (0.20 ± 0.21 mm) but the differences did not reach statistical significance (P = 0.40). The acute percent recoil demonstrated the same trend (EES: 4.3% ± 7.1%, BVS 1.0: 6.9% ± 7.0%, BVS 1.1: 6.7% ± 6.4%, P = 0.22). In the multivariate regression model, high balloon/artery ratio (>1.1) (OR 1.91 [1.34-2.71]) was the predictive for high absolute recoil (>0.27 mm) while (larger) preprocedural MLD was protective (OR 0.84 [0.72-0.99]). The stent/scaffold type was not a predictor of acute recoil. CONCLUSIONS: The average in vivo acute scaffold recoil of the BVS 1.1 is slightly higher than the metallic EES. However, the scaffold/stent type was not predictive of high acute recoil, while implantation in undersized vessels or usage of oversized devices might confound the results.

Automatic generation of minimum dataset and quality indicators from data collected routinely by the clinical information system in an intensive care unit.

BACKGROUND: Quality indicators (QIs) are being increasingly used in medicine to compare and improve the quality of care delivered. The feasibility of data collection is an important prerequisite for QIs. Information technology can improve efforts to measure processes and outcomes. In intensive care units (ICU), QIs can be automatically measured by exploiting data from clinical information systems (CIS). OBJECTIVE: To describe the development and application of a tool to automatically generate a minimum dataset (MDS) and a set of ICU quality metrics from CIS data. METHODS: We used the definitions for MDS and QIs proposed by the Spanish Society of Critical Care Medicine and Coronary Units. Our tool uses an extraction, transform, and load process implemented with Python to extract data stored in various tables in the CIS database and create a new associative database. This new database is uploaded to Qlik Sense, which constructs the MDS and calculates the QIs by applying the required metrics. The tool was tested using data from patients attended in a 30-bed polyvalent ICU during a six-year period. RESULTS: We describe the definitions and metrics, and we report the MDS and QI measurements obtained through the analysis of 4546 admissions. The results show that our ICU's performance on the QIs analyzed meets the standards proposed by our national scientific society. CONCLUSIONS: This is the first step toward using a tool to automatically obtain a set of actionable QIs to monitor and improve the quality of care in ICUs, eliminating the need for professionals to enter data manually, thus saving time and ensuring data quality.

A Machine Learning decision-making tool for extubation in Intensive Care Unit patients.

BACKGROUND AND OBJECTIVE: To increase the success rate of invasive mechanical ventilation weaning in critically ill patients using Machine Learning models capable of accurately predicting the outcome of programmed extubations. METHODS: The study population was adult patients admitted to the Intensive Care Unit. Target events were programmed extubations, both successful and failed. The working dataset is assembled by combining heterogeneous data including time series from Clinical Information Systems, patient demographics, medical records and respiratory event logs. Three classification learners have been compared: Logistic Discriminant Analysis, Gradient Boosting Method and Support Vector Machines. Standard methodologies have been used for preprocessing, hyperparameter tuning and resampling. RESULTS: The Support Vector Machine classifier is found to correctly predict the outcome of an extubation with a 94.6% accuracy. Contrary to current decision-making criteria for extubation based on Spontaneous Breathing Trials, the classifier predictors only require monitor data, medical entry records and patient demographics. CONCLUSIONS: Machine Learning-based tools have been found to accurately predict the extubation outcome in critical patients with invasive mechanical ventilation. The use of this important predictive capability to assess the extubation decision could potentially reduce the rate of extubation failure, currently at 9%. With about 40% of critically ill patients eventually receiving invasive mechanical ventilation during their stay and given the serious potential complications associated to reintubation, the excellent predictive ability of the model presented here suggests that Machine Learning techniques could significantly improve the clinical outcomes of critical patients.

Understanding critically ill sepsis patients with normal serum lactate levels: results from U.S. and European ICU cohorts.

While serum lactate level is a predictor of poor clinical outcomes among critically ill patients with sepsis, many have normal serum lactate. A better understanding of this discordance may help differentiate sepsis phenotypes and offer clues to sepsis pathophysiology. Three intensive care unit datasets were utilized. Adult sepsis patients in the highest quartile of illness severity scores were identified. Logistic regression, random forests, and partial least square models were built for each data set. Features differentiating patients with normal/high serum lactate on day 1 were reported. To exclude that differences between the groups were due to potential confounding by pre-resuscitation hyperlactatemia, the analyses were repeated for day 2. Of 4861 patients included, 47% had normal lactate levels. Patients with normal serum lactate levels had lower 28-day mortality rates than those with high lactate levels (17% versus 40%) despite comparable physiologic phenotypes. While performance varied between datasets, logistic regression consistently performed best (area under the receiver operator curve 87-99%). The variables most strongly associated with normal serum lactate were serum bicarbonate, chloride, and pulmonary disease, while serum sodium, AST and liver disease were associated with high serum lactate. Future studies should confirm these findings and establish the underlying pathophysiological mechanisms, thus disentangling association and causation.

Varying association of laboratory values with reference ranges and outcomes in critically ill patients: an analysis of data from five databases in four countries across Asia, Europe and North America.

BACKGROUND: Despite wide usage across all areas of medicine, it is uncertain how useful standard reference ranges of laboratory values are for critically ill patients. OBJECTIVES: The aim of this study is to assess the distributions of standard laboratory measurements in more than 330 selected intensive care units (ICUs) across the USA, Amsterdam, Beijing and Tarragona; compare differences and similarities across different geographical locations and evaluate how they may be associated with differences in length of stay (LOS) and mortality in the ICU. METHODS: A multi-centre, retrospective, cross-sectional study of data from five databases for adult patients first admitted to an ICU between 2001 and 2019 was conducted. The included databases contained patient-level data regarding demographics, interventions, clinical outcomes and laboratory results. Kernel density estimation functions were applied to the distributions of laboratory tests, and the overlapping coefficient and Cohen standardised mean difference were used to quantify differences in these distributions. RESULTS: The 259 382 patients studied across five databases in four countries showed a high degree of heterogeneity with regard to demographics, case mix, interventions and outcomes. A high level of divergence in the studied laboratory results (creatinine, haemoglobin, lactate, sodium) from the locally used reference ranges was observed, even when stratified by outcome. CONCLUSION: Standardised reference ranges have limited relevance to ICU patients across a range of geographies. The development of context-specific reference ranges, especially as it relates to clinical outcomes like LOS and mortality, may be more useful to clinicians.

Prognostic Value of Procalcitonin and C-Reactive Protein in 1608 Critically Ill Patients with Severe Influenza Pneumonia.

Background: Procalcitonin (PCT) and C-Reactive protein (CRP) are well-established sepsis biomarkers. The association of baseline PCT levels and mortality in pneumonia remains unclear, and we still do not know whether biomarkers levels could be related to the causative microorganism (GPC, GNB). The objective of this study is to address these issues. Methods: a retrospective observational cohort study was conducted in 184 Spanish ICUs (2009-2018). Results: 1608 patients with severe influenza pneumonia with PCT and CRP available levels on admission were included, 1186 with primary viral pneumonia (PVP) and 422 with bacterial Co-infection (BC). Those with BC presented higher PCT levels (4.25 [0.6-19.5] versus 0.6 [0.2-2.3]ng/mL) and CRP (36.7 [20.23-118] versus 28.05 [13.3-109]mg/dL) as compared to PVP (p < 0.001). Deceased patients had higher PCT (ng/mL) when compared with survivors, in PVP (0.82 [0.3-2.8]) versus 0.53 [0.19-2.1], p = 0.001) and BC (6.9 [0.93-28.5] versus 3.8 [0.5-17.37], p = 0.039). However, no significant association with mortality was observed in the multivariate analysis. The PCT levels (ng/mL) were significantly higher in polymicrobial infection (8.4) and GPC (6.9) when compared with GNB (1.2) and Aspergillus (1.7). The AUC-ROC of PCT for GPC was 0.67 and 0.32 for GNB. The AUROC of CRP was 0.56 for GPC and 0.39 for GNB. Conclusions: a single PCT/CRP value at ICU admission was not associated with mortality in severe influenza pneumonia. None of the biomarkers have enough discriminatory power to be used for predicting the causative microorganism of the co-infection.