Lieuwe D. Bos

Volatile Metabolites of Pathogens: A Systematic Review

Ideally, invading bacteria are detected as early as possible in critically ill patients: the strain of morbific pathogens is identified rapidly, and antimicrobial sensitivity is known well before the start of new antimicrobial therapy. Bacteria have a distinct metabolism, part of which results in the production of bacteria-specific volatile organic compounds (VOCs), which might be used for diagnostic purposes. Volatile metabolites can be investigated directly in exhaled air, allowing for noninvasive monitoring. The aim of this review is to provide an overview of VOCs produced by the six most abundant and pathogenic bacteria in sepsis, including Staphylococcus aureus, Streptococcus pneumoniae, Enterococcus faecalis, Pseudomonas aeruginosa, Klebsiella pneumoniae, and Escherichia coli. Such VOCs could be used as biological markers in the diagnostic approach of critically ill patients. A systematic review of existing literature revealed 31 articles. All six bacteria of interest produce isopentanol, formaldehyde, methyl mercaptan, and trimethylamine. Since humans do not produce these VOCs, they could serve as biological markers for presence of these pathogens. The following volatile biomarkers were found for identification of specific strains: isovaleric acid and 2-methyl-butanal for Staphylococcus aureus; 1-undecene, 2,4-dimethyl-1-heptane, 2-butanone, 4-methyl-quinazoline, hydrogen cyanide, and methyl thiocyanide for Pseudomonas aeruginosa; and methanol, pentanol, ethyl acetate, and indole for Escherichia coli. Notably, several factors that may effect VOC production were not controlled for, including used culture media, bacterial growth phase, and genomic variation within bacterial strains. In conclusion, VOCs produced by bacteria may serve as biological markers for their presence. Goal-targeted studies should be performed to identify potential sets of volatile biological markers and evaluate the diagnostic accuracy of these markers in critically ill patients.

Interobserver agreement of Centers for Disease Control and Prevention criteria for classifying infections in critically ill patients

Objectives:Correct classification of the source of infection is important in observational and interventional studies of sepsis. Centers for Disease Control and Prevention criteria are most commonly used for this purpose, but the robustness of these definitions in critically ill patients is not known. We hypothesized that in a mixed ICU population, the performance of these criteria would be generally reduced and would vary among diagnostic subgroups. Design:Prospective cohort. Setting:Data were collected as part of a cohort of 1,214 critically ill patients admitted to two hospitals in The Netherlands between January 2011 and June 2011. Patients:Eight observers assessed a random sample of 168 of 554 patients who had experienced at least one infectious episode in the ICU. Each patient was assessed by two randomly selected observers who independently scored the source of infection (by affected organ system or site), the plausibility of infection (rated as none, possible, probable, or definite), and the most likely causative pathogen. Assessments were based on a post hoc review of all available clinical, radiological, and microbiological evidence. The observed diagnostic agreement for source of infection was classified as partial (i.e., matching on organ system or site) or complete (i.e., matching on specific diagnostic terms), for plausibility as partial (2-point scale) or complete (4-point scale), and for causative pathogens as an approximate or exact pathogen match. Interobserver agreement was expressed as a concordant percentage and as a kappa statistic. Interventions:None. Measurements and Main Results:A total of 206 infectious episodes were observed. Agreement regarding the source of infection was 89% (183/206) and 69% (142/206) for a partial and complete diagnostic match, respectively. This resulted in a kappa of 0.85 (95% CI, 0.79–0.90). Agreement varied from 63% to 91% within major diagnostic categories and from 35% to 97% within specific diagnostic subgroups, with the lowest concordance observed in cases of ventilator-associated pneumonia. In the 142 episodes for which a complete match on source of infection was obtained, the interobserver agreement for plausibility of infection was 83% and 65% on a 2- and 4-point scale, respectively. For causative pathogen, agreement was 78% and 70% for an approximate and exact pathogen match, respectively. Conclusions:Interobserver agreement for classifying sources of infection using Centers for Disease Control and Prevention criteria was excellent overall. However, full concordance on all aspects of the diagnosis between independent observers was rare for some types of infection, in particular for ventilator-associated pneumonia.

Exhaled Molecular Fingerprinting in Diagnosis and Monitoring: Validating Volatile Promises

Medical diagnosis and phenotyping increasingly incorporate information from complex biological samples. This has promoted the development and clinical application of non-invasive metabolomics in exhaled air (breathomics). In respiratory medicine, expired volatile organic compounds (VOCs) are associated with inflammatory, oxidative, microbial, and neoplastic processes. After recent proof of concept studies demonstrating moderate to good diagnostic accuracies, the latest efforts in breathomics are focused on optimization of sensor technologies and analytical algorithms, as well as on independent validation of clinical classification and prediction. Current research strategies are revealing the underlying pathophysiological pathways as well as clinically-acceptable levels of diagnostic accuracy. Implementing recent guidelines on validating molecular signatures in medicine will enhance the clinical potential of breathomics and the development of point-of-care technologies.

A European Respiratory Society technical standard: exhaled biomarkers in lung disease

Breath tests cover the fraction of nitric oxide in expired gas (FENO), volatile organic compounds (VOCs), variables in exhaled breath condensate (EBC) and other measurements. For EBC and for FENO, official recommendations for standardised procedures are more than 10 years old and there is none for exhaled VOCs and particles. The aim of this document is to provide technical standards and recommendations for sample collection and analytic approaches and to highlight future research priorities in the field. For EBC and FENO, new developments and advances in technology have been evaluated in the current document. This report is not intended to provide clinical guidance on disease diagnosis and management. Clinicians and researchers with expertise in exhaled biomarkers were invited to participate. Published studies regarding methodology of breath tests were selected, discussed and evaluated in a consensus-based manner by the Task Force members. Recommendations for standardisation of sampling, analysing and reporting of data and suggestions for research to cover gaps in the evidence have been created and summarised. Application of breath biomarker measurement in a standardised manner will provide comparable results, thereby facilitating the potential use of these biomarkers in clinical practice. ERS technical standard: exhaled biomarkers in lung disease http://ow.ly/mAjr309DBOP

Exhaled breath metabolomics as a noninvasive diagnostic tool for acute respiratory distress syndrome

There is a need for biological markers of the acute respiratory distress syndrome (ARDS). Exhaled breath contains hundreds of metabolites in the gas phase, some of which reflect (patho)physiological processes. We aimed to determine the diagnostic accuracy of metabolites in exhaled breath as biomarkers of ARDS. Breath from ventilated intensive care unit patients (n=101) was analysed using gas chromatography and mass spectrometry during the first day of admission. ARDS was defined by the Berlin definition. Training and temporal validation cohorts were used. 23 patients in the training cohort (n=53) had ARDS. Three breath metabolites, octane, acetaldehyde and 3-methylheptane, could discriminate between ARDS and controls with an area under the receiver operating characteristic curve (AUC) of 0.80. Temporal external validation (19 ARDS cases in a cohort of 48) resulted in an AUC of 0.78. Discrimination was insensitive to adjustment for severity of disease, a direct or indirect cause of ARDS, comorbidities, or ventilator settings. Combination with the lung injury prediction score increased the AUC to 0.91 and improved net reclassification by 1.17. Exhaled breath analysis showed good diagnostic accuracy for ARDS, which was externally validated. These data suggest that exhaled breath analysis could be used for the diagnostic assessment of ARDS. Metabolites in the breath of ventilated patients may be used to diagnose the acute respiratory distress syndrome http://ow.ly/uWHF1

Classification of patients with sepsis according to blood genomic endotype: a prospective cohort study

BACKGROUND Host responses during sepsis are highly heterogeneous, which hampers the identification of patients at high risk of mortality and their selection for targeted therapies. In this study, we aimed to identify biologically relevant molecular endotypes in patients with sepsis. METHODS This was a prospective observational cohort study that included consecutive patients admitted for sepsis to two intensive care units (ICUs) in the Netherlands between Jan 1, 2011, and July 20, 2012 (discovery and first validation cohorts) and patients admitted with sepsis due to community-acquired pneumonia to 29 ICUs in the UK (second validation cohort). We generated genome-wide blood gene expression profiles from admission samples and analysed them by unsupervised consensus clustering and machine learning. The primary objective of this study was to establish endotypes for patients with sepsis, and assess the association of these endotypes with clinical traits and survival outcomes. We also established candidate biomarkers for the endotypes to allow identification of patient endotypes in clinical practice. FINDINGS The discovery cohort had 306 patients, the first validation cohort had 216, and the second validation cohort had 265 patients. Four molecular endotypes for sepsis, designated Mars1-4, were identified in the discovery cohort, and were associated with 28-day mortality (log-rank p=0·022). In the discovery cohort, the worst outcome was found for patients classified as having a Mars1 endotype, and at 28 days, 35 (39%) of 90 people with a Mars1 endotype had died (hazard ratio [HR] vs all other endotypes 1·86 [95% CI 1·21-2·86]; p=0·0045), compared with 23 (22%) of 105 people with a Mars2 endotype (HR 0·64 [0·40-1·04]; p=0·061), 16 (23%) of 71 people with a Mars3 endotype (HR 0·71 [0·41-1·22]; p=0·19), and 13 (33%) of 40 patients with a Mars4 endotype (HR 1·13 [0·63-2·04]; p=0·69). Analysis of the net reclassification improvement using a combined clinical and endotype model significantly improved risk prediction to 0·33 (0·09-0·58; p=0·008). A 140-gene expression signature reliably stratified patients with sepsis to the four endotypes in both the first and second validation cohorts. Only Mars1 was consistently significantly associated with 28-day mortality across the cohorts. To facilitate possible clinical use, a biomarker was derived for each endotype; BPGM and TAP2 reliably identified patients with a Mars1 endotype. INTERPRETATION This study provides a method for the molecular classification of patients with sepsis to four different endotypes upon ICU admission. Detection of sepsis endotypes might assist in providing personalised patient management and in selection for trials. FUNDING Center for Translational Molecular Medicine, Netherlands.

Breathomics in the setting of asthma and chronic obstructive pulmonary disease

Exhaled breath contains thousands of volatile organic compounds that reflect the metabolic process occurring in the host both locally in the airways and systemically. They also arise from the environment and airway microbiome. Comprehensive analysis of breath volatile organic compounds (breathomics) provides opportunities for noninvasive biomarker discovery and novel mechanistic insights. Applications in patients with obstructive lung diseases, such as asthma and chronic obstructive pulmonary disease, include not only diagnostics (especially in children and other challenging diagnostic areas) but also identification of clinical treatable traits, such as airway eosinophilia and risk of infection/exacerbation, that are not specific to diagnostic labels. Although many aspects of breath sampling and analysis are challenging, proof-of-concept studies with mass spectrometry and electronic nose technologies have provided independent studies with moderate-to-good diagnostic and phenotypic accuracies. The present review evaluates the data obtained by using breathomics in (1) predicting the inception of asthma or chronic obstructive pulmonary disease, (2) inflammatory phenotyping, (3) exacerbation prediction, and (4) treatment stratification. The current findings merit the current efforts of large multicenter studies using standardized sampling, shared analytic methods, and databases, including external validation cohorts. This will position this noninvasive technology in the clinical assessment and monitoring of chronic airways diseases.

The volatile metabolic fingerprint of ventilator-associated pneumonia

Dear Editor, The diagnostic approach for ventilator-associated pneumonia (VAP) needs to be improved [1]. Volatile organic compounds (VOCs), produced either by invading respiratory pathogens or the patient’s pulmonary defense system, could serve as early diagnostic markers for VAP [2]. Electronic nose (eNose) technology integratively captures complex VOC mixtures to create a ‘VOC fingerprint’ using an array of semi-selective sensors [3]. We hypothesized that an eNose would be able to discriminate patients with VAP from those without VAP based on analysis of headspace air from tracheal aspirates (TAs). In a prospective cohort study we collected TAs every third day from 45 intensive care unit (ICU) patients who were ventilated for more than 7 days. Fourteen patients developed VAP, 14 patients had airway colonization but did not develop VAP and 17 patients developed neither VAP nor airway colonization (study methodology and patient characteristics are given in the Electronic Supplementary Material). The eNose was able to accurately discriminate patients with VAP from those without VAP in both a cross-sectional and longitudinal analysis (Fig. 1, upper panels), and the use of a ‘VOC fingerprint’ was found to improve the diagnostic accuracy of the Clinical Pulmonary Infection Score (Fig. 1, lower panels) in this small cohort of patients. Notably, discrimination by the eNose was not affected by airway colonization, and the findings were independent of the number of colony forming units in the TAs. Our study has several limitations. First, we were not able to identify which VOCs differentiate between patients with VAP and those without VAP. Furthermore, this study was performed in a highly selected cohort of patients, and the sample size was rather small, thereby limiting generalization of our findings. Indeed, the results of our study need to be confirmed in robust and larger studies. Of interest, the results of our study suggest that the observed changes in VOC-fingerprints are not solely the result of the presence or absence of bacteria in TAs. VOC-fingerprints can change with the bacterial ecology from colonization to infection, which

Identification and validation of distinct biological phenotypes in patients with acute respiratory distress syndrome by cluster analysis

Rationale We hypothesised that patients with acute respiratory distress syndrome (ARDS) can be clustered based on concentrations of plasma biomarkers and that the thereby identified biological phenotypes are associated with mortality. Methods Consecutive patients with ARDS were included in this prospective observational cohort study. Cluster analysis of 20 biomarkers of inflammation, coagulation and endothelial activation provided the phenotypes in a training cohort, not taking any outcome data into account. Logistic regression with backward selection was used to select the most predictive biomarkers, and these predicted phenotypes were validated in a separate cohort. Multivariable logistic regression was used to quantify the independent association with mortality. Results Two phenotypes were identified in 454 patients, which we named ‘uninflamed’ (N=218) and ‘reactive’ (N=236). A selection of four biomarkers (interleukin-6, interferon gamma, angiopoietin 1/2 and plasminogen activator inhibitor-1) could be used to accurately predict the phenotype in the training cohort (area under the receiver operating characteristics curve: 0.98, 95% CI 0.97 to 0.99). Mortality rates were 15.6% and 36.4% (p<0.001) in the training cohort and 13.6% and 37.5% (p<0.001) in the validation cohort (N=207). The ‘reactive phenotype’ was independent from confounders associated with intensive care unit mortality (training cohort: OR 1.13, 95% CI 1.04 to 1.23; validation cohort: OR 1.18, 95% CI 1.06 to 1.31). Conclusions Patients with ARDS can be clustered into two biological phenotypes, with different mortality rates. Four biomarkers can be used to predict the phenotype with high accuracy. The phenotypes were very similar to those found in cohorts derived from randomised controlled trials, and these results may improve patient selection for future clinical trials targeting host response in patients with ARDS.

Comparison of classification methods in breath analysis by electronic nose

Currently, many different methods are being used for pre-processing, statistical analysis and validation of data obtained by electronic nose technology from exhaled air. These various methods, however, have never been thoroughly compared. We aimed to empirically evaluate and compare the influence of different dimension reduction, classification and validation methods found in published studies on the diagnostic performance in several datasets. Our objective was to facilitate the selection of appropriate statistical methods and to support reviewers in this research area. We reviewed the literature by searching Pubmed up to the end of 2014 for all human studies using an electronic nose and methodological quality was assessed using the QUADAS-2 tool tailored to our review. Forty-six studies were evaluated regarding the range of different approaches to dimension reduction, classification and validation. From forty-six reviewed articles only seven applied external validation in an independent dataset, mostly with a case-control design. We asked their authors to share the original datasets with us. Four of the seven datasets were available for re-analysis. Published statistical methods for eNose signal analysis found in the literature review were applied to the training set of each dataset. The performance (area under the receiver operating characteristics curve (ROC-AUC)) was calculated for the training cohort (in-set) and after internal validation (leave-one-out cross validation). The methods were also applied to the external validation set to assess the external validity of the performance. Risk of bias was high in most studies due to non-random selection of patients. Internal validation resulted in a decrease in ROC-AUCs compared to in-set performance:  -0.15,-0.14,-0.1,-0.11 in dataset 1 through 4, respectively. External validation resulted in lower ROC-AUC compared to internal validation in dataset 1 (-0.23) and 3 (-0.09). ROC-AUCs did not decrease in dataset 2 (+0.07) and 4 (+0.04). No single combination of dimension reduction and classification methods gave consistent results between internal and external validation sets in this sample of four datasets. This empirical evaluation showed that it is not meaningful to estimate the diagnostic performance on a training set alone, even after internal validation. Therefore, we recommend the inclusion of an external validation set in all future eNose projects in medicine.