[Differential diagnosis in fever of unknown origin: significance of concomitant clinical symptoms]

Between 1978 and 1984, 169 patients were admitted to the hospital for fever of unknown origin which was repeatedly above 38.3 degrees C. After a retrospective analysis of their records the patients were divided into two groups on the basis of the following new criteria. The first group (74 patients) was described as having "monosymptomatic fever", i.e. fever without any other physical signs, whereas the second group (95 patients) had "polysymptomatic fever", i.e. fever with additional physical signs. In 56 patients (76%) of the monosymptomatic group fever had lasted longer than 3 weeks prior to admission. In 86% of these patients case history, physical examination, microbiological tests, serological tests for microorganisms and outoimune antibodies, and microscopic inspections of tissue and/or bone marrow led to a diagnosis. Malignancies, factitious fever and fever of unknown origin were found only in this group. The patients with malignancies were generally older than the rest of the patients (p less than 0.05), and eight of ten patients suffering from connective tissue diseases also had monosymptomatic fever. The incidence of infections in this group was 42% (31 cases), in contrast to 88% (84 cases) in the polysymptomatic group (p less than 0.05). Whereas the latter had significantly more bacterial infections (p less than 0.05), viral infections prevailed in the monosymptomatic group (p less than 0.05). Thus, the etiology of polysymptomatic fever distinctly differed from that of monosymptomatic fever. Since the frequency distribution of etiologies in the monosymptomatic group corresponded to that of the cases of fever of unknown origin in the literature, differentiation into monosymptomatic and polysymptomatic fever might be helpful in determining further diagnostic workup of patients with fever of unknown origin.