No abstract

Nonpigmented multiply drug-resistant Serratia marcescens caused an extensive outbreak of infection at the Nashville Veterans Administration Hospital. Isolates were of one serotype resistant to all currently available antimicrobial agents for therapy of systemic infections except for occasional susceptibility to chloramphenicol and kanamycin. Frequently strains were susceptible to nalidixic acid, and all were susceptible to amikacin (BB-K8). Drug-resistant strains caused 130 infections, 12 bacteremias, and 7 infection-associated deaths. Combinations of antimicrobial agents were evaluated for synergism against Serratia strains from infected patients. “Checkerboard” isobolograms indicated in vitro static synergism between sulfamethoxazole, trimethoprim, and polymyxin (STP). Killing curves using clinically achievable concentrations of STP verified the bactericidal effect of STP against these strains. In a daily dosage of 1,600 mg of sulfamethoxazole and 320 mg of trimethoprim orally in combination with 100 to 300 mg of colistimethate parenterally, serum cidal levels at 1:8 or greater were achieved in five of six patients. Clinical improvement or microbiological cure was effected in four of six patients. STP may be potentially useful for selected Serratia infections for which single agents are unavailable or ineffective.

SUMMARY Fifty eight patients undergoing dental extraction each had 45 ml blood collected. This was divided into 30 ml and 15 ml blood samples for culture. The 30 ml sample was inoculated into 120 ml nutrient broth with 0-05% liquoid and the 15 ml sample into 60 ml of identical broth so that the final dilution of blood in broth was always 1/5.Bacteraemia due to viridans streptococci was found in 27 and 15 patients by culturing the 30 ml and 15 ml blood samples respectively. Only one further case of streptococcal bacteraemia was detected by culture of the total volume of blood collected (45 ml) rather than culture of the 30 ml blood sample alone. These findings suggest that the culture of 30 ml blood results in the detection of up to 80% more blood cultures yielding Streptococcus viridans than the culture of only 15 ml blood. The collection of more than 30 ml blood for each culture is unlikely to prove worthwhile. It is suggested that 30 ml rather than 15 ml blood is probably the optimal volume of blood for each culture of S, viridans when patients with suspected infective endocarditis are investigated.Only a few reported studies have determined the independent effect of volume of blood cultures on the detection of bacteraemia where other variables such as the ratio of blood to broth, the medium, atmosphere of incubation, and processing methods were identical for all blood cultures.' 3 A recent study at the Mayo Clinic of bacteraemia indicated that the yield of 15 ml samples is 25% greater than that of 5 ml samples.4 In the Association of Clinical Pathologists Broadsheet a 15 ml blood sample is recommended for culture.5 None of the reports published so far have commented on bacteraemia due to viridans streptococci. Streptococcus viridans is still the most common cause of endocarditis, and patients often have a low grade bacteraemia. Although the culture of a large volume of blood would be expected to yield more streptococci than the culture of a small volume, there are no reports indicating the optimal volume of blood for the isolation of viridans streptococci. There are too few patients with streptococcal endocarditis to allow worthwhile studies to be carried out. Patients undergoing dental extractions, however, may also have predictable S viridans bacteraemias. We report the results of a study on patients undergoing dental extraction, where the results of culture of 30 ml and 15 ml blood samples were compared and all Accepted for publication 2 February 1984 the other blood culture factors were controlled so that blood volume was the only variable.

No abstract

No abstract