Are incidence and epidemiology of anaerobic bacteremia really changing?

Incidence, prognosis and need of performing blood cultures for anaerobic bacteria are under debate, mainly due to the belief that the presence of anaerobes in blood can be easily suspected on clinical basis. We aimed to assess these three points in a retrospective analysis of a 10-year experience in our tertiary hospital. All episodes of significant anaerobic bacteremia diagnosed from 2003 to 2012 were included. Risk factors for mortality and clinical predictability of anaerobic bacteremia were evaluated in 113 randomly selected episodes. Overall incidence of anaerobic bacteremia was 1.2 episodes/1000 admissions, with no significant changes during the 10-year study period. B. fragilis group (38.1 %) and Clostridium spp. (13.7 %) were the most frequent isolated microorganisms. As for the clinical study, 43.4 % of the patients had a comorbidity classified as ultimately fatal or rapidly fatal according to the McCabe and Jackson scale. Clinical manifestations suggestive of anaerobic involvement were present in only 55 % of the patients. Twenty-eight patients (24.8 %) died during the hospitalization. Independent predictive factors of mortality were a high Charlson's comorbidity index and presentation with septic shock, whereas, an adequate source control of the infection was associated with a better outcome. In our centre, incidence of anaerobic bacteremia remained stable during the last decade. The routine use of anaerobic BCs seems to be adequate, since in about half of the cases anaerobes could not be suspected on clinical bases. Moreover, prompt source control of infection is essential in order to reduce mortality of patients with anaerobic bacteremia.