Early Biomarker Signatures in Surgical Sepsis

Abstract

Introduction: Sepsis has complex, time-sensitive pathophysiology and important phenotypic subgroups. The objective of this study was to use machine learning analyses of blood and urine biomarker profiles to elucidate the pathophysiologic signatures of subgroups of surgical sepsis patients.

Methods: This prospective cohort study included 243 surgical sepsis patients admitted to a quaternary care center between January 2015 and June 2017. We applied hierarchical clustering to clinical variables and 42 blood and urine biomarkers to identify phenotypic subgroups in a development cohort. Clinical characteristics and short-term and long-term outcomes were compared between clusters. A naïve Bayes classifier predicted cluster labels in a validation cohort.

Results: The development cohort contained one cluster characterized by early organ dysfunction (cluster I, n = 18) and one cluster characterized by recovery (cluster II, n = 139). Cluster I was associated with higher Acute Physiologic Assessment and Chronic Health Evaluation II (30 versus 16, P < 0.001) and SOFA scores (13 versus 5, P < 0.001), greater prevalence of chronic cardiovascular and renal disease (P < 0.001) and septic shock (78% versus 17%, P < 0.001). Cluster I had higher mortality within 14 d of sepsis onset (11% versus 1.5%, P = 0.001) and within 1 y (44% versus 20%, P = 0.032), and higher incidence of chronic critical illness (61% versus 30%, P = 0.001). The Bayes classifier achieved 95% accuracy and identified two clusters that were similar to development cohort clusters.

Conclusions: Machine learning analyses of clinical and biomarker variables identified an early organ dysfunction sepsis phenotype characterized by inflammation, renal dysfunction, endotheliopathy, and immunosuppression, as well as poor short-term and long-term clinical outcomes.

Keywords: Biomarker; Clustering; Machine learning; Phenotyping; Sepsis; Unsupervised learning.

Fig. 1 –

Patient enrollment and inclusion flowchart.

Fig. 2 –

Clustering dendrogram with significance. (A) Dendrogram showing the subject profile arrangement from the hierarchical clustering using the complete linkage method with normalized distances by the number of features 42. (B) Cluster composition for different levels of significance α. Colors track clusters that are robust with increasing significance levels. (C) QQ-plot displaying the observed and expected distances used in hierarchical clustering (height of branch nodes). Departure from the diagonal line suggests that there are significant clusters in the data.

Fig. 3 –

Distributions of biomarkers by clusters. Side-by-side boxplots show the distributions of the standardized biomarker values across clusters. Each color represents a group of biomarkers based on their functionality. In each plot, the horizontal dash line represents the average value of the standardized biomarker in the whole cohort.

Fig. 4 –

Average biomarker mosaics for patients using a self-organizing map for each of the two main clusters. Average biomarker mosaics of a specific cluster illustrate the average value of biomarker values for patients within that cluster. Red color correlates with increased biomarker expression, and blue color correlates with decreased biomarker expression compared to mean values for the development cohort, which is illustrated by the green color.