Matt Harmon

Chapter four 77 Introduction Body temperature changes are common in sepsis. 1 Patients who present with spontaneous hypothermia suffer from substantially increased morbidity and mortality compared to their normothermic or febrile counterparts. 2,3 It is unclear whether hypothermia simply represents a symptom of severe inflammation or that hypothermia itself drives mortality through a yet unknown mechanism. 4,5 Animal studies even indicate that hypothermia may be an adaptive response to severe inflammation in order to limit metabolism and prevent hypoxia. 6 To determine the etiology of the hypothermic response during sepsis, studies have mainly focused on the ability to generate an adequate host immune response, often with levels of pro-inflammatory cytokines as a read-out. However, studies have not confirmed a defective immune host response, as septic patients with hypothermia had similar 7,8 , or even increased proinflammatory cytokine levels 9 in comparison to normothermic or febrile patients. Patients with hypothermic sepsis do develop persistent lymphopenia, a marker of immunosuppression. 10 Alternatively, the cardiovascular system may play a role. We previously showed that markers of endothelial injury are increased in hypothermic sepsis compared to nonhypothermic controls. 7 Taken together, the pathophysiology of hypothermic response in sepsis remains ill defined. Whole blood transcriptome analysis has provided valuable insights in the complex pathophysiology of the sepsis syndrome. 11 In this study we aimed to determine if hypothermia in sepsis patients is associated with a different blood leukocyte gene expression profile compared to fever. We hypothesized that the blood transcriptomes of hypothermic sepsis patients differed from those obtained in febrile sepsis patients, which in turn reflect on variations in the host response. Materials and methods Study Design, Setting, and Patient Identification This study was performed within the ‘Molecular Diagnosis and Risk Stratification of Sepsis (MARS) project, a prospective observational cohort study in mixed ICUs of two tertiary teaching hospitals (Academic Medical Center in Amsterdam and University Medical Center in Utrecht) in the Netherlands (ClinicalTrials.gov identifier NCT01905033). 12,13 Between January 2011 and July 2012, patients older than 18 years of age with an expected length of stay longer than 24 hours were included via an opt-out consent method approved by ethical committees of both hospitals (IRB no. 10-056). During this study demographic, clinical, microbiology,