LAUSR

The relationship between disease severity, immunosuppression, and outcome The immunological response in sepsis patients is complex, comprising concurrent proand anti-inflammatory responses. It has become increasingly clear that most sepsis patients do not succumb to an early, overwhelming pro-inflammatory response, but rather to complications related to immunosuppression occurring later on in their disease trajectory [1, 2]. This severely suppressed state of the immune system renders patients unable to clear their primary infection and increases susceptibility toward secondary infections, often with opportunistic pathogens [3, 4]. Recently, a seminal observational study questioned the contribution of secondary infections to sepsis mortality [5]. This study linked disease severity in sepsis patients to increased susceptibility toward secondary infections and higher mortality. Analysis of the transcriptome of circulating leukocytes revealed distinct hallmarks of immune suppression at the onset of secondary infections, consistent with an important role for sepsis-induced immunosuppression as a causative factor. However, after adjustment for age and, crucially, disease severity the attributable mortality from secondary infections was low (2%) [5]. This may appear to indicate that sepsis-induced immunosuppression is of limited importance for outcome, and that interventions aimed to mitigate or reverse it will therefore have little impact. We believe this is not the case, because immunosuppression is an intermediary factor in sepsis patients. Disease severity drives immunosuppression, eventually leading to mortality related to secondary infections. It is therefore not surprising that correcting for disease severity abolishes the contribution of secondary infections to mortality. This does, however, not mean that therapeutic interventions aimed at an intermediary factor, in this case immunosuppression, may not affect outcome. We would like to illustrate this using diabetes as an example. In diabetic patients, poor glycemic control (indicated by increased HbA1c) induces vascular damage, resulting in an increased rate of myocardial infarctions. In this example, HbA1c reflects disease severity, vascular damage the intermediary factor, and myocardial infarction the outcome. If the attributable myocardial infarctions of vascular damage would be corrected for HbA1c in diabetic patients, the effect would be modest as well. However, it is widely accepted that vascular damage is causative for myocardial infarctions. More importantly, interventions targeting vascular damage (e.g., percutaneous coronary intervention) are highly effective, also in patients with diabetes. Analogously, strategies aimed at mitigating sepsis-induced immunosuppression should not be written off.