Thesis Judith Zandstra

Summary

For the proper treatment of patients , the right diagnosis is essential. Yet, getting to the correct diagnosis is often challenging when specific indicators are lacking that may help the treating physician. Fever is considered to be one of the key symptoms of infection that can be caused by bacteria, viruses, yeast or fungi. To overcome this limitation and to get a better understanding of the cause of the fever, we have set out to identify specific plasma protein patterns in the patients ’ blood to aid the physician to make a more probable diagnosis. In this thesis we show the strength of combining several plasma protein biomarkers to diagnose the source of fever in different group of patients . In heterogeneous discovery cohorts we measured the plasma levels of various protein biomarkers and performed independent validations in separate cohorts. We are one step closer to developing a more sensitive test to improve early diagnosis and treatment practices by distinguishing bacterial from viral infections as well as infection from inflammatory disease. Despite the progress made, our studies in selected cohorts also illustrate that additional real-world application in different settings and study populations should be tested. We should also be aware of the fact that a predictive model may never be final and complete, nor globally applicable since any diagnostic biomarker model built for rapid testing will depend on the selection of biomarkers , based on local exposure rates of most prevalent infections, newly emerging infections during epidemics, host genetics, and many other variables of regional relevance.