Prior attempts at identifying outcome determinants associated with bloodstream infection have employed a priori determined classification schemes based on readily identifiable microbiology, infection site, and patient characteristics. We hypothesized that even amongst this heterogeneous population, clinically relevant groupings can be described that transcend old a priori classifications.
Cluster analysis was applied to variables from three domains: patient characteristics, acuity of illness/clinical presentation and infection characteristics.
Among 3715 patients with bloodstream infections from Barnes-Jewish Hospital (2008 to 2015), the most stable cluster arrangement occurred with the formation of four clusters. This clustering arrangement resulted in an approximately uniform distribution of the population: Clusters One (21.5%), Two (27.9%), Three (28.7%) and Four (21.9%). Staphylococcus aureus distributed primarily to Clusters Three (40%) and Four (25%), while Enterobacteriaceae were divided predominantly into Clusters Two (34%), Three (30%), and Four (22%). Nonfermenting Gram-negative bacilli grouped mainly in Clusters Four and Two (30% and 31%). More than half of the pneumonia cases occurred in Cluster Four. Clusters One and Two contained 33% and 31% respectively of the individuals receiving inappropriate antibiotic administration. Mortality was greatest for Cluster Four (33.8%, 27.4%, 19.2%, 44.6%; P < 0.001), while Cluster One patients were most likely to be discharged to a nursing home.
Our results support the potential for machine learning methods to identify homogenous groupings in infectious diseases that transcend old a priori classifications. These methods may allow new clinical phenotypes to be identified potentially improving the severity staging and treatment of complex infectious diseases.
M. C. Vazquez Guillamet,
S. Micek, None
M. Kollef, None