Teaching fever aphorisms: Osler revisited

Abstract

Sir William Osler was a master clinician, renowned for his clinical method, attention to detail, and his diagnostic acumen. In Osler’s era, infectious diseases were rampant and Osler had the opportunity to observe and write about many infectious diseases that he encountered at the time. In his famous Textbook of Medicine, published in 1892, pride of place goes to infectious diseases, and the infectious disease section begins with 39 pages devoted to typhoid fever. Because diagnostic tests available at the time were limited, Osler based his diagnoses on a careful analysis of the patient’s febrile response and correlated that with the clinical findings to arrive at a diagnosis. Osler was able to correctly diagnose infectious diseases with similar presenting symptoms based on his careful analysis of the patients’ fever curves. He was also careful to note the relationship of each patient’s pulse to fever [1, 2]. In the 1850s, the most important differential diagnostic problem in infectious disease was the differentiation of typhoid fever from malaria. Most clinicians at the time were so confused by the similarity of symptoms of these disorders that the diagnosis of “typhomalaria” was used. “Typhomalaria” was a testimony to the lack of attention being paid to the details of fever curves associated with typhoid fever compared to those of malaria [3–5]. Osler not only appreciated that relative bradycardia was a cardinal sign of typhoid fever, he also understood the temporal relationships between relative bradycardia and the natural history of the infection. In the medical literature today, many authors report that relative bradycardia is not often present in typhoid fever and therefore is a less reliable diagnostic finding than previously described [1, 6]. Osler was careful to note that relative bradycardia occurred late rather than early in the natural history of typhoid fever. Today, we rarely have the opportunity to observe the untreated natural course of an infectious disease for more than a few days because of economic constraints on length of stay for hospitalized patients. However, this does not detract from the validity of his observations and it accounts for the discrepancy between his and contemporary descriptions of the incidence and diagnostic importance of relative bradycardia in typhoid fever [7, 8]. Osler solved the “typhomalarial” diagnostic dilemma by carefully comparing the fever curves and pulse relationships of these two different infectious diseases, thereby debunking “typhomalaria” forever. He also noted that coinfection is rare; patients have either typhoid fever or malaria and rarely, if ever, both simultaneously [1, 8]. Clinicians today often lament the excesses and undue reliance on diagnostic testing at the expense of careful diagnostic methods at the bedside. The lack of careful observation and diagnostic reasoning skills all too often lacking in some house officers is regrettable but, in part, it is the fault of many training programs. More time is spent ordering CT scans than in analyzing the patient’s fever curves and correlating them with clinical findings. How many hours of house staff core lectures or conferences throughout the academic year are devoted to the diagnostic significance of fever? If included in core lectures, fever of unknown origin, or the pathophysiology of fever are the topics covered most often. House officers are faced with diagnosing acute fevers in their hospitalized patients every Eur J Clin Microbiol Infect Dis (2007) 26:371–373 DOI 10.1007/s10096-007-0286-4