Prevalence of Gardnerella vaginalis in the urinary tract.

Midstream urine samples from 106 patients presenting to the Casualty Department of The Royal Melbourne Hospital with frequency or dysuria were cultured for Gardnerella vaginalis and conventional uropathogens. Urine samples collected via an open-end catheter from 70 healthy pregnant women were examined similarly. Midstream urine and other samples, including the seminal fluids and swabs of the mouths, throats, rectums, and vaginas of 33 healthy subjects, were cultured for G. vaginalis. Another 15 female patients with proven G. vaginalis bacteriuria were given a bladder washout localization test to determine the site of infection in the urinary tract. G. vaginalis in counts greater than 10(3) CFU/ml was recovered from the midstream urine of 27 of 106 patients (25%), 7 of whom also harbored conventional pathogens in counts greater than 10(4) CFU/ml. Another 11 patients with cultures negative for G. vaginalis yielded greater than 10(4) CFU of conventional pathogens per ml. G. vaginalis was cultured (greater than 10(3) CFU/ml) from catheter samples of 19 of 70 healthy pregnant women (27%), 6 of whom also harbored greater than 10(3) CFU of conventional uropathogens per ml. Two women yielded growths of conventional pathogens only. Midstream urine samples from 13 of 13 healthy males were free of G. vaginalis, whereas 5 of 20 healthy nonpregnant females yielded greater than 10(3) CFU of G. vaginalis per ml from midstream urine samples. G. vaginalis was recovered from 4 of 12 semen samples and from urethral samples from four of seven males and four of eight females. All four culture-positive females also harbored G. vaginalis in their vaginas. There was no evidence of oral or rectal carriage of G. vaginalis in 15 healthy subjects. Localization studies with 15 female patients having underlying renal disease showed that 11 patients harbored G. vaginalis in their kidneys. The result suggest that colonization or infection of the bladder and upper urinary tract by G. vaginalis is very largely a phenomenon of females, with the highest frequency in pregnant women. The prevalence of G. vaginalis in the urinary tracts of healthy females is similar to that of symptomatic subjects. However, G. vaginalis in counts greater than 10(5) CFU/ml is more likely to be associated with urinary tract symptoms. In males, this bacterial species infects the genital tract rather than the urinary tract.