Combination therapy and extending treatment for 4 weeks or longer gave significantly better results than monotherapy or shorter courses of therapy and resulted in fewer relapses.

In a comparison of the Bactec system and the lysis concentration procedure in the isolation of Brucella species in 54 patients the recovery rate was similar (60% and 55%, respectively). However, the recovery time was significantly shorter with the lysis concentration method than with the Bactec system (3.5 days versus 14 days). The lysis concentration procedure for the culture of Brucella is simple, inexpensive and reliable, and produces results for the clinician relatively quickly.

The infection rate (percentage of mice shedding 1 organisms per ml of urine) in 27 mice infected intravesicularly with a mannose-specific (MS') phenotype of KkbsieUa pneumoniae was 85% at day 7, and all the bacteria shed during the 7 days exhibited strong MS activity as estimated by a yeast aggregation assay. In contrast, the outcome of infection with an MSphenotype of the same strain in 47 mice was heterogenous: one group of 25 mice continued to shed the originally injected phenotype (MS-) throughout the investigation period, whereas the second group (22 mice) shed bacteria with various degrees of phenotypic conversion to MS'. In the first group, the rate of infection at day 7 was significantly reduced (28%) compared with that of the second group (68%). Mice infected with a mixture of 5% MS' bacteria and 95% of an MS-variant which lost its ability to undergo phase variation had an infection rate of 89%, but at day 7 95% of the excreted bacteria were MS'. The infection rate of mice injected with the MS-variant was 14%, and none of the mice shed MS' bacteria. The incidence of kidney pathology was higher in mice inoculated with the MS' phenotype (3 of 10) or in the group in which the MS' overgrew the MS-phenotype (4 of 10) as compared with the group of mice in which no such shift occurred (1 of 11). The kidneys of four mice which excreted mostly MS' organisms harbored a population predominantly of the MS-phenotype. These results suggest that the MS adhesin confers an advantage in the initial steps of the infectious process in the bladder but not in later stages of infection in the kidney, emphasizing the importance of phase variation in the survival of bacteria at the various stages of the infectious process.

The role of Escherichia coli isolates from the semen in the etiology of male infertility, was investigated in this study. Several possible virulence factors of E. coli, such as possession of 0 antigens or certain K antigens (K,, K5), the type of fimbriae, resistance to antimicrobial drugs and adherence studies to various mammalian sperm cells were examined. It was found that out of 181 E. coli isolates 76.4% belonged to four different urinary serotypes: 0,, 0,, 0, and 0, (5.7y0, 7.5y0, 17.0% and 46.2y0, respectively). The predominant fimbrial phenotype (81.2%) was T l F + / P -, while 15.8% showed also the possession of P fimbriae ( T l F + / P + ) . No isolate was found with the T l F -/ P+ phenotype. The possession of antigens K,, K, was found to be low (12% only). A relatively high level of adherence to different mammalian sperm cells was found among E. coli isolates. 43% of the E. coli isolates were resistant to doxycycline (the drug of choice) and additionally, most of these were resistant also to ampicillin. We suggest that there is a close subpopulation of E. coli which possess certain virulence properties and have the potential to adhere to sperm cells and to colonize on other target tissues in the male genital tract. Such virulent strains may cause asymptomatic male infertility and may be termed male genital tract (MGT)-E. coli.