Preclinical studies have suggested that platelets influence the host response during sepsis. We sought to assess the association of admission thrombocytopenia with the presentation, outcome, and host response in patients with sepsis. Nine hundred thirty-one consecutive sepsis patients were stratified according to platelet counts (very low <50 × 10⁹/L, intermediate-low 50 × 10⁹ to 99 × 10⁹/L, low 100 × 10⁹ to 149 × 10⁹/L, or normal 150 × 10⁹ to 399 × 10⁹/L) on admission to the intensive care unit. Sepsis patients with platelet counts <50 × 10⁹/L and 50 × 10⁹ to 99 × 10⁹/L presented with higher Acute Physiology and Chronic Health Evaluation scores and more shock. Both levels of thrombocytopenia were independently associated with increased 30-day mortality (hazard ratios with 95% confidence intervals 2.00 [1.32-3.05] and 1.72 [1.22-2.44], respectively). To account for baseline differences besides platelet counts, propensity matching was performed, after which the association between thrombocytopenia and the host response was tested, as evaluated by measuring 17 plasma biomarkers indicative of activation and/or dysregulation of pathways implicated in sepsis pathogenesis and by whole genome blood leukocyte expression profiling. In the propensity matched cohort, platelet counts < 50 × 10⁹/L were associated with increased cytokine levels and enhanced endothelial cell activation. All thrombocytopenic groups showed evidence of impaired vascular integrity, whereas coagulation activation was similar between groups. Blood microarray analysis revealed a distinct gene expression pattern in sepsis patients with <50 × 10⁹/L platelets, showing reduced signaling in leukocyte adhesion and diapedesis and increased complement signaling. These data show that admission thrombocytopenia is associated with enhanced mortality and a more disturbed host response during sepsis independent of disease severity, thereby providing clinical validity to animal studies on the role of platelets in severe infection. Publisher9s Note: There is an Inside Blood Commentary on this article in this issue. Key Points Thrombocytopenia on intensive care unit admission is independently associated with increased mortality in patients with sepsis. Thrombocytopenia is associated with a more disturbed host response in critically ill patients with sepsis independent of disease severity.

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.

Rationale: Community-acquired pneumonia (CAP) accounts for a major proportion of intensive care unit (ICU) admissions for respiratory failure and sepsis. Diagnostic uncertainty complicates case management, which may delay appropriate cause-specific treatment. Objectives: To characterize the blood genomic response in patients with suspected CAP and identify a candidate biomarker for the rapid diagnosis of CAP on ICU admission. Methods: The study comprised two cohorts of consecutively enrolled patients treated for suspected CAP on ICU admission. Patients were designated CAP (cases) and no-CAP patients (control subjects) by post hoc assessment. The first (discovery) cohort (101 CAP and 33 no-CAP patients) was enrolled between January 2011 and July 2012; the second (validation) cohort (70 CAP and 30 no-CAP patients) between July 2012 and June 2013. Blood was collected within 24 hours of ICU admission. Measurements and Main Results: Blood microarray analysis of CAP and no-CAP patients revealed shared and distinct gene expression patterns. A 78-gene signature was defined for CAP, from which a FAIM3:PLAC8 gene expression ratio was derived with area under curve of 0.845 (95% confidence interval, 0.764–0.917) and positive and negative predictive values of 83% and 81%, respectively. Robustness of the FAIM3:PLAC8 ratio was ascertained by quantitative polymerase chain reaction in the validation cohort. The FAIM3:PLAC8 ratio outperformed plasma procalcitonin and IL-8 and IL-6 in discriminating between CAP and no-CAP patients. Conclusions: CAP and no-CAP patients presented shared and distinct blood genomic responses. We propose the FAIM3:PLAC8 ratio as a candidate biomarker to assist in the rapid diagnosis of CAP on ICU admission. Clinical trial registered with www.clinicaltrials.gov (NCT 01905033).

Introduction: A clinical suspicion of infection is mandatory for diagnosing sepsis in patients with a systemic inflammatory response syndrome. Yet, the accuracy of categorizing critically ill patients presenting to the intensive care unit (ICU) as being infected or not is unknown. We therefore assessed the likelihood of infection in patients who were treated for sepsis upon admission to the ICU, and quantified the association between plausibility of infection and mortality. Methods: We studied a cohort of critically ill patients admitted with clinically suspected sepsis to two tertiary ICUs in the Netherlands between January 2011 and December 2013. The likelihood of infection was categorized as none, possible, probable or definite by post-hoc assessment. We used multivariable competing risks survival analyses to determine the association of the plausibility of infection with mortality. Results: Among 2579 patients treated for sepsis, 13% had a post-hoc infection likelihood of “none”, and an additional 30% of only “possible”. These percentages were largely similar for different suspected sites of infection. In crude analyses, the likelihood of infection was associated with increased length of stay and complications. In multivariable analysis, patients with an unlikely infection had a higher mortality rate compared to patients with a definite infection (subdistribution hazard ratio 1.23; 95% confidence interval 1.03-1.49). Conclusions: This study is the first prospective analysis to show that the clinical diagnosis of sepsis upon ICU admission corresponds poorly with the presence of infection on post-hoc assessment. A higher likelihood of infection does not adversely influence outcome in this population. Trial registration: ClinicalTrials.gov NCT01905033. Registered 11 July 2013.

Rationale: Patients admitted to intensive care units with sepsis are prone to developing cardiac dysrhythmias, most commonly atrial fibrillation. Objectives: To determine the incidence, risk factors, and outcomes of atrial fibrillation in a cohort of critically ill patients with sepsis. Methods: We assessed the association between atrial fibrillation and mortality using time-dependent competing risks survival analysis. Subsequently, for development of a risk score estimating the probability of a first occurrence of atrial fibrillation within the following 24 hours, we performed logistic regression analysis. Measurements and Main Results: Among 1,782 patients with sepsis admitted to two tertiary intensive care units in the Netherlands between January 2011 and June 2013, a total of 1,087 episodes of atrial fibrillation occurred in 418 (23%) individuals. The cumulative risk of new-onset atrial fibrillation was 10% (95% confidence interval [CI], 8–12), 22% (95% CI, 18–25), and 40% (95% CI, 36–44) in patients with sepsis, severe sepsis, and septic shock, respectively. New-onset atrial fibrillation was associated with a longer stay (hazard ratio [HR], 0.55; 95% CI, 0.48–0.64), an increased death rate (HR, 1.52; 95% CI, 1.16–2.00), and an overall increased mortality risk (subdistribution HR, 2.10; 95% CI, 1.61–2.73) when considering discharge as a competing event. A simple risk score for daily prediction of atrial fibrillation occurrence yielded good discrimination (C statistic, 0.81; 95% CI, 0.79–0.84) and calibration (chi-square, 9.38; P = 0.31), with similar performance in an independent validation cohort (C statistic, 0.80; 95% CI, 0.76–0.85). Conclusions: Atrial fibrillation is a common complication of sepsis and independently associated with excess mortality. A simple risk score may identify patients at high risk of this complication. Clinical trial registered with www.clinicaltrials.gov (NCT 01905033).