[Utility of anaerobic blood cultures in extra-hospitalary bacteremias]

To compare the rentability of anaerobic hemocultures with aerobic cultures in patients suffering bacteremia, trying to analyze the rentability of anaerobic cultures.

There were analyzed all the hemocultures taken by medical decision for one month. There were taken three extractions by patient in different venopunctures points. In each extraction was taken 10 ml (5 ml in aerobic bottle, and 5 ml in anaerobic bottle). There were compared microorganism isolated in aerobic bottles and anaerobic bottles. Three physician analyzed all the clinical dates of the patients with positive hemocultures and classified them as negatives, contaminants and significant.

There were taken hemocultures of 180 patients. There were analyzed 152 hemocultures groups (3 extractions aerobic-anaerobic), without studying 28 owe to not following the study protocol 24 (13.3%) or because of not possibility of studying clinical story 4 (2.2%). Aerobic extractions: Negatives 91 (59.86%), contaminants 36 (23.68%), significant with or without contaminant 25 (16.44%). Anaerobic extractions: Negatives 127 (83.5%), contaminants 2 (1.31%); significant with or without contaminant 23 (15.13%). There are 6 (8.6%) significant positive isolated in anaerobic bottles, that were negative in aerobic bottles, and are: 3 E. coli, 1 K. pneumoniae, 1 Peptostreptococcus sp, 1 Bacteroides sp, and over the total significant isolated (31) are the 19.35%. Both anaerobic bacteremias were clinically suspected. Contaminant microorganism more frequently isolated in aerobic bottles was Staphylococcus sp, 25 cases (69.44% of all contaminants). Significant microorganism more frequently isolated was E. coli (in anaerobic bottles too) in 12 cases (38.70%) of all significant isolated, following by Streptococcus pneumoniae with 4 cases (12.9%).

We recommend collection of one aerobic and one anaerobic blood culture bottle per blood culture set because the global rentability of hemocultures is increased significantly.