A study of the susceptibility of three species of primate to vaginal colonization with Gardnerella vaginalis

In an attempt to develop an animal model of Gardnerella-associated vaginitis, several strains of Gardnerella vaginalis were inoculated into the lower genital tract of female pig-tailed macaques, tamarins and chimpanzees. G. vaginalis was not recovered from either tamarins or chimpanzees, but was recovered from each of 1O pig-tailed macaques inoculated with either of two freshly isolated Gardnerella strains, colonization persisting for 11-39 days. Examination of Gram-stained vaginal smears obtained from infected pig-tailed macaques failed to demonstrate clue cells, a feature which is pathognomonic of Gardnerella-associated vaginitis in humans. Other features characteristic of non-specific vaginitis, namely an increase in vaginal pH, and an increase in the ratio of succinate to lactate (S/L ratio) in vaginal fluid were not found. However, the physiology of the macaque vagina was found to be different from that of the human, the vaginal pH and S/L ratio of uninfected macaques both being higher than that seen in humans. The physiological differences between the macaque and human vagina may be due, in part, to a difference in their anaerobic vaginal flora. While these inter-species differences in vaginal physiology and microbiology limit the relevance of the pig-tailed macaque as a model of Gardnerella-associated vaginitis, the ease with which macaques are colonized with G. vaginalis may prove useful in studying bacterial adhesion and local immunity.     

Comparison of culture and microscopy in the diagnosis of Gardnerella vaginalis infection.

A comparison was made between human blood agar containing amphotericin B, nalidixic acid and either gentamicin or colistin for the isolation of Gardnerella vaginalis from cases of non-specific vaginitis seen in a clinic for sexually transmitted diseases. The medium containing gentamicin was more inhibitory for non-Gardnerella species, but not sufficiently inhibitory to allow direct plating in the clinic without spreading for single colonies. The diffuse beta haemolysis produced by G vaginalis on human, but not on horse blood agar, proved very useful in differentiating it from other vaginal organisms and was not affected by the antibiotics used. This characteristic, together with Gram stain morphology, oxidase and catalase, provides a simple, reliable methods of identifying G vaginalis. Sixty women with symptoms of vaginitis, in whom no other pathogen was isolated, were examined by culture and microscopy. Gardnerella vaginalis was grown from 45 whereas only 31 had positive microscopy (clue cells or Gram-variable bacilli). There was no significant difference between the rate of isolation of G vaginalis in the group with positive microscopy (25/31) and that with negative microscopy (20/31).     

Selective differential human blood bilayer media for isolation of Gardnerella (Haemophilus) vaginalis

New selective and differential human blood bilayer agar media with Tween 80 (HBT medium) or without Tween 80 (HB medium), developed for the isolation of Gardnerella (Haemophilus) vaginalis, permitted significantly higher G. vaginalis isolation rates than have been obtained for other media used for this purpose. HB medium consists of a basal layer of Columbia agar base containing colistin and naladixic acid with added amphotericin B and an overlayer of the same composition plus 5% human blood. HBT agar also contains Proteose Peptone No. 3 (Difco Laboratories) and Tween 80 in the basal layer and the overlayer. Both Tween 80 and the bilayer composition enhanced G. vaginalis production of human blood hemolysis, permitting detection of this organism even in the presence of heavy growth of other vaginal flora. The use of HB or HBT medium thus permitted the demonstration that G. vaginalis was present in vaginal fluid from a large percentage (up to 68%) of normal women. However, the concentration of G. vaginalis was found by semiquantitative analysis to be significantly higher in vaginal fluid from women with nonspecific vaginitis than in fluid from normal women.     

Biotypes of Gardnerella vaginalis.

A simple and reproducible scheme for identifying biotypes of Gardnerella vaginalis has been developed, based on reactions for lipase, hippurate hydrolysis, and beta-galactosidase. Among a total of 359 strains tested, eight biotypes were observed, the most common ones being types 1 (beta-galactosidase positive, lipase positive, hippurate positive), 2 (beta-galactosidase negative, lipase positive, hippurate positive), and 5 (beta-galactosidase negative, lipase negative, hippurate positive). The distribution in biotypes was similar among isolates from Antwerp, Seattle, and Nairobi. There were no differences in biotypes between strains isolated from patients with and without bacterial vaginosis (nonspecific vaginitis). Up to 14% of women with bacterial vaginosis harbored at least two different biotypes of G. vaginalis in the vagina. G. vaginalis strains isolated before and after treatment for bacterial vaginosis belonged to identical biotypes when the time interval between two specimens was less than 1 week. Similarly, G. vaginalis isolates from the vaginas of women with bacterial vaginosis and from the urethras of their male sex partners belonged to identical biotypes when strains were isolated within the same 24-h period from both partners (P less than 0.005).     

Possible role of enterotoxigenic Bacteroides fragilis in the etiology of infectious vaginitis

Vaginitis is a common gynecologic disorder. It is due to several causes, some even unknown. Bacteroides fragilis is the most important anaerobe in clinical bacteriology, some strains of this group are notable for being enterotoxigenic and they have been associated with intestinal and extraintestinal syndromes. They have recently been isolated from patients with vaginitis. The purpose of this study was to investigate a possible association of enterotoxigenic B. fragilis with infectious vaginitis. 265 samples of vaginal exudate were processed, 202 from symptomatic patients and 63 healthy women. The identification of the microorganisms was carried out by conventional methods. In 31.2% of symptomatic patients were identified: Gardnerella vaginalis, Mobiluncus, Candida albicans, Mycoplasma hominis, Ureaplasma urealyticum and Streptococcus agalactiae. B. fragilis was identified in 27 symptomatic patients and 5 healthy women. These strains were cultivated in liquid medium and incubated during 48 h at 36 degrees C in anaerobe chambers. Supernatant activity was assayed in HT-29 cells. Eighteen B. fragilis strains isolated from symptomatic patients were enterotoxigenic, because induced alterations in target cell morphology. It was not identified in healthy women (P < 0.05). 77.7% of enterotoxigenic B. fragilis strains were not associated with other specific pathogens. This fact suggests that enterotoxigenic B. fragilis could be a cause for vaginitis. The effect of enterotoxin on E-cadherin of vaginal epithelium could facilitate invasion and its possible pathogenic role in the vagina. This is the first report that associates enterotoxigenic Bacteroides fragilis as a possible cause of infectious vaginitis.     

Semiquantitative culture of Gardnerella vaginalis in laboratory determination of nonspecific vaginitis

To evaluate the usefulness of quantitative cultures of Gardnerella vaginalis in the laboratory determination of nonspecific vaginitis, the actual and relative numbers of G. vaginalis in genital cultures of a general patient population were assessed semiquantitatively, and the laboratory results were then correlated with the clinical findings. Of the 1,585 women studied, 417 (26.3%) yielded G. vaginalis in culture. Of these, only 113 (27.1%) were found to have symptoms and signs consistent with nonspecific vaginitis. G. vaginalis was obtained in pure or predominant growth from 87 of 100 consecutive cases with nonspecific vaginitis and 32 of 100 consecutive cases without the symptoms or signs of vaginitis (P less than 0.001). Hence, the positive predictive value of isolation of G. vaginalis in pure and predominant growths was determined to be 73% (87 of 119). Conversely, G. vaginalis was isolated in mixed or light growth significantly more often from asymptomatic women than from symptomatic patients, i.e., 68 versus 13 cases. Therefore, the negative predictive value of isolation of G. vaginalis in mixed and light growths was found to be 84% (68 of 81). Quantitation of the relative amount of G. vaginalis growth had higher predictive values as compared with the assessment of G. vaginalis growth alone. We conclude that quantitative culture of G. vaginalis is essential to obtain maximum reliability of culture results in the laboratory determination of nonspecific vaginitis. Although quantitated cultures of G. vaginalis have high predictive values, laboratory results must be interpreted in conjunction with the clinical findings.     

Mixed vaginal infections of Balb/c mice with low virulent herpes simplex type 1 strains result in restoration of virulence properties: vaginitis/vulvitis and neuroinvasiveness

Vaginal infections of BALB/c Ann mice with herpes simplex virus type 1 (HSV-1) were studied. Mice were inoculated with virulent strains ANG path and 17 syn⁺ or low-virulent recombinant strains 27/III and 17-syn3 that differ from parental strains in their glycoprotein B (gB) gene sequences. When low-virulent strains were inoculated separately, no vaginitis/vulvitis was produced despite replication in the vagina. In contrast, after coinfection of mice with the two low-virulent strains, vaginitis/vulvitis was produced and virus could be recovered from the central nervous system (CNS). Two of the CNS isolates produced vaginitis/vulvitis, neuroinvasiveness and death of mice after vaginal infection. Restriction fragment analysis and sequencing were used to assess recombination events in the gB gene sequence of the CNS isolates. After mixed vaginal infection recombination between non-virulent HSV strains occurs, resulting in vaginitis/vulvitis and neuroinvasiveness. No correlation was detected between the syncytial phenotype and local vaginal virulence. Virulence of HSV is not solely dependent on gB function; it seems to be more probable that several genes act in concert to induce virulence and neuroivasiveness. Received: 28 May 1996     

Gardnerella vaginalis: characteristics, clinical considerations, and controversies.

The clinical significance, Gram stain reaction, and genus affiliation of Gardnerella vaginalis have been controversial since Gardner and Dukes described the organism as the cause of "nonspecific vaginitis," a common disease of women which is now called bacterial vaginosis. The organism was named G. vaginalis when taxonomic studies showed that it was unrelated to bacteria in various genera including Haemophilus and Corynebacterium. Electron microscopy and chemical analyses have elucidated the organism's gram-variable reaction. Controversy over the etiology of bacterial vaginosis was largely resolved by (i) studies using improved media and methods for the isolation and identification of bacteria in vaginal fluids and (ii) standardization of criteria for clinical and laboratory diagnosis. Besides G. vaginalis, Mobiluncus spp., Mycoplasma hominis, and certain obligate anaerobes are now acknowledged as participants in bacterial vaginosis. The finding that G. vaginalis, Mobiluncus spp., and M. hominis inhabit the rectum indicates a potential source of autoinfection in addition to sexual transmission. Extravaginal infections with G. vaginalis are increasingly recognized, especially when the toxic anticoagulant polyanetholesulfonate is omitted from blood cultures and when urine cultures are incubated anaerobically for 48 h. The finding that mares harbor G. vaginalis suggests that an equine model can be developed for studies of Gardnerella pathogenesis.     

Non-volatile fatty acids in the diagnosis of non-specific vaginitis.

In the vaginal washings of 100 women with symptomatic non-specific vaginitis a succinate/lactate ratio of greater than or equal to 0.4 had a diagnostic sensitivity of 80%, a specificity of 83% for this condition. The predictive value of a positive test was 94%, but that of a negative test was only 55%. A strong association between the presence of Gardnerella vaginalis, anaerobes, a vaginal pH of above 4.5, and amines was found not only in non-specific vaginitis, but also in trichomonal and gonococcal infection. A variety of primary changes may encourage the multiplication of both gardnerellae and anaerobes and their presence in non-specific vaginitis may be a secondary rather than a primary event.     

Longitudinal study of the biotypes of Gardnerella vaginalis.

Gardnerella vaginalis is the predominant vaginal microorganism in women with bacterial vaginosis. However, this organism is also frequently isolated from women without signs or symptoms of vaginitis. Earlier studies have not revealed whether certain biotypes of G. vaginalis are more often associated with bacterial vaginosis or are more common in women who acquire bacterial vaginosis. We used a typing scheme based on tests for beta-galactosidase, hippurate hydrolysis, and lipase, using oleate as a substrate. Of 261 strains tested, the distribution of biotypes observed was as follows: 1, 13%; 2, 9%; 3, 5%; 4, 7%; 5, 41%; 6, 15%; and 8, 10%. Biotype 7 was not observed. The distributions of biotypes from women with and without bacterial vaginosis were found to be significantly different, with the lipase-positive biotypes (biotypes 1, 2, 3, and 4) being more predominant in women with vaginosis (41 versus 23%, P = 0.003). Of 40 women with normal vaginal flora at the index visit who remained normal at follow-up, 23 (57%) acquired a new biotype of G. vaginalis. By comparison, 90% of the 30 women who developed bacterial vaginosis acquired a new biotype of G. vaginalis (P = 0.003). Women with bacterial vaginosis at the index visit who were not treated were no more likely than normal women to have a shift in G. vaginalis biotype. However, 86% of the 30 women with bacterial vaginosis who were treated with an antibiotic at the index visit acquired a different biotype (P = 0.04 compared with the value for untreated women) regardless of treatment success. A trend toward the acquisition of a new biotype was observed among women who had contact with a new sexual partner (81 versus 65%, P = 0.15). These data demonstrate that the lipase-positive isolates of G. vaginalis are associated with bacterial vaginosis. Women who acquire bacterial vaginosis are more likely to have a shift in biotype than women who had normal flora at he follow-up, suggesting that the G. vaginalis isolates recovered from women who develop bacterial vaginosis represent newly acquired strains rather than overgrowth of previously colonizing biotypes.     

Isolation and identification of Gardnerella vaginalis

Thirty-four strains of Gardnerella vaginalis were studied. They were isolated from non specific vaginitis. A presumptive identification can be based on colonial morphology, Gram stain characteristics, negative catalase and oxidase test. The differentiation of Gardnerella vaginalis from other negative catalase coccobacilli is based on the acid formation of carbohydrates, enzymatic test, analysis of short chain volatile and non volatile end products of fermentation in GLC. All Gardnerella strains produce acid from glucose, maltose, starch and are hippurate and alpha glucosidase positive. They are sulfonamide resistant, acetic, lactic and succinic acids are detected by gas liquid chromatography.     

The presence of the casein kinase II phosphorylation sites of Vpu enhances the CD4(+) T cell loss caused by the simian-human immunodeficiency virus SHIV(KU-lbMC33) in pig-tailed macaques

The simian-human immunodeficiency virus (SHIV)/ macaque model for human immunodeficiency virus type 1 has become a useful tool to assess the role of Vpu in lentivirus pathogenesis. In this report, we have mutated the two phosphorylated serine residues of the HIV-1 Vpu to glycine residues and have reconstructed a SHIV expressing this nonphosphorylated Vpu (SHIV(S52,56G)). Expression studies revealed that this protein was localized to the same intracellular compartment as wild-type Vpu. To determine if this virus was pathogenic, four pig-tailed macaques were inoculated with SHIV(S52,56G) and virus burdens and circulating CD4(+) T cells monitored up to 1 year. Our results indicate that SHIV(S52,56G) caused rapid loss in the circulating CD4(+) T cells within 3 weeks of inoculation in one macaque (CC8X), while the other three macaques developed no or gradual numbers of CD4(+) T cells and a wasting syndrome. Histological examination of tissues revealed that macaque CC8X had lesions in lymphoid tissues (spleen, lymph nodes, and thymus) that were typical for macaques inoculated with pathogenic parental SHIV(KU-1bMC33) and had no lesions within the CNS. To rule out that macaque CC8X had selected for a virus in which there was reversion of the glycine residues at positions 52 and 56 to serine residues and/or compensating mutations occurred in other genes associated with CD4 down-regulation, sequence analysis was performed on amplified vpu sequences isolated from PBMC and from several lymphoid tissues at necropsy. Sequence analysis revealed a reversion of the glycine residues back to serine residues in this macaque. The other macaques maintained low virus burdens, with one macaque (P003) developing a wasting syndrome between months 9 and 11. Histological examination of tissues from this macaque revealed a thymus with severe atrophy that was similar to that of a previously reported macaque inoculated with a SHIV lacking vpu (Virology 293, 2002, 252). Sequence analysis revealed no reversion of the glycine residues in the vpu sequences isolated from this macaque. These results contrast with those from four macaques inoculated with the parental pathogenic SHIV(KU-1bMC33), all of which developed severe CD4(+) T cell loss within 1 month after inoculation. Taken together, these results indicate that casein kinase II phosphorylation sites of Vpu contributes to the pathogenicity of the SHIV(KU-1bMC33) and suggest that the SHIV(KU-1bMC33)/pig-tailed macaque model will be useful in analyzing amino acids/domains of Vpu that contribute to the pathogenesis of HIV-1.     

Nonspecific vaginitis: role of Haemophilus vaginalis and treatment with metronidazole.

To assess the cause of nonspecific vaginitis, we performed a prospective case-control study of vaginal flora and a randomized unblinded trial of different therapies. Haemophilus vaginalis was isolated from 17 to 18 women with signs of vaginitis but only one of 18 normal matched controls (P less than 0.002). The concentration of anaerobic bacteria in vaginal washings also was increased in patients. Clinical improvement and eradication of H. vaginalis occurred in one of seven patients given sulfonamide vaginal cream, two of 15 given oral doxycycline, nine of 27 given oral ampicillin, and 80 of 81 given oral metronidazole. On the seventh day of therapy signs of nonspecific vaginitis persisted in 31 of 31 with, and in two of 92 without, persistent H. vaginalis infection (P less than 0.001). These data suggest the causal role of H. vaginalis in nonspecific vaginitis, possibly in concert with vaginal anaerobes. The widespread use of sulfonamide creams is inappropriate. Metronidazole is effective, but its efficacy must be weighed against its possible toxicity.     

Effect of sodium polyanetholesulfonate and gelatin on the recovery of Gardnerella vaginalis from blood culture media.

Sodium polyanetholesulfonate (SPS) is used as a routine supplement to blood culture media to enhance recovery of microorganisms, but it inhibits the growth of Peptostreptococcus anaerobius, Neisseria meningitidis, Neisseria gonorrhoeae, and Streptobacillus moniliformis. Comparative clinical blood culture studies at the University of Colorado Hospital suggested that SPS also inhibits the growth of Gardnerella vaginalis. We inoculated 16 blood culture isolates of G. vaginalis into 11 blood culture media containing SPS or sodium amylosulfate, with and without gelatin. In the absence of gelatin, only brain heart infusion and thiol broths with SPS supported the growth of more than five strains of G. vaginalis, whereas all media except Bactec 6B and 7C and brucella broths recovered most isolates with SPS and gelatin or with sodium amylosulfate alone. We conclude that SPS inhibits the growth of G. vaginalis in blood culture media but that this inhibition is medium dependent and can be overcome by supplementation of most media with gelatin.     

Biotypes and virulence factors of Gardnerella vaginalis isolated from cases of bacterial vaginosis

The present study was conducted to correlate the biotypes of Gardnerella vaginalis strains isolated from cases of bacterial vaginosis and their virulence factors. Thirty-two strains of G. vaginalis isolated from cases of bacterial vaginosis were biotyped. Adherence to vaginal epithelial cells, biofilm production, surface hydrophobicity, phospholipase C and protease activity were tested on these isolates. Biotype 1 was the most prevalent (8; 25%), followed by biotype 2 (7; 21.9%) and biotypes 5 and 8 (5; 15.6%). We did not find any statistical correlation between G. vaginalis biotypes and its virulence factors. Virulence factors expressed by G. vaginalis were not associated with a single biotype.     

Humoral circulatory immune response to Gardnerella vaginalis.

Strain-specific circulating immunoglobulin G and/or M was detected by enzyme-linked immunosorbent assay and immunofluorescence test by using Formol-treated suspensions of Gardnerella vaginalis from 28 women with overt vaginitis but only three symptom-free subjects among 43 otherwise healthy women found to be colonized by G. vaginalis. Analogous but less stringent strain specificity patterns were elicited by immunization of BALB/c mice.     

Characterization of Gardnerella vaginalis and G. vaginalis-like organisms from the reproductive tract of the mare.

Gardnerella vaginalis has been isolated from women with bacterial vaginosis, from the genital tracts of asymptomatic women, and from several other infected body sites in humans. However, until recently, it has not been isolated from any other animal species. Between June 1988 and October 1989, 31 isolates identified as G. vaginalis and 70 isolates identified as G. vaginalis-like organisms have been recovered from the genital tracts of 93 mares from Michigan and Ohio. Identification was based on biochemical reactions, hemolysis on media containing blood from various animal sources, and susceptibility to select antimicrobial agents. This report details the characterization of G. vaginalis and G. vaginalis-like organism isolates obtained from the reproductive tracts of these mares and compares the equine isolates with human isolates.     

A rotavirus strain isolated from pig-tailed macaque (Macaca nemestrina) with diarrhea bears a P6[1]:G8 specificity

A distinct rotavirus strain (PTRV) was isolated in cell cultures from a stool sample obtained from a diarrheic 3-year-old female pig-tailed macaque (Macaca nemestrina) that was born at the breeding colony of the University of Washington in Seattle. Unlike other known simian rotavirus strains including vervet monkey rotavirus SA11 which bears P5B[2]:G3 or P6[1]:G3 specificity, rhesus monkey rotavirus MMU18006 with P5B[3]:G3 specificity, pig-tailed macaque rotavirus YK-1 with P[3]:G3 specificity and rhesus monkey rotavirus TUCH with P[24]:G3 specificity, the cell-culture-grown PTRV strain was shown to bear P6[1]:G8 specificity as determined by VP4 (P)- and VP7 (G)-specific neutralization assays as well as gene sequence analyses. The virus in the original diarrhea stool was also shown to bear genotypes P[1] and G8. In addition, the PTRV strain exhibited a "long" electropherotype, subgroup I specificity and NSP4 genotype A specificity. The PTRV probe formed (i) 8-9 hybrid bands with genomic RNAs of various bovine rotavirus strains and (ii) only 2-3 hybrid bands with simian rotavirus RNAs as demonstrated by RNA-RNA hybridization, suggesting a possible bovine origin of the virus. Serologic analysis of serum samples obtained from selected pig-tailed macaques in the colony suggested that a rotavirus bearing P[1]:G8 specificity was endemic among macaques for at least 8 years (1987-1994). This is the first report describing an isolation of a simian rotavirus bearing a non-G3 VP7 and possibly a P6[1] specificities. Because of its unique simian serotype, this virus may prove to be valuable in challenge studies in a non-human primate model in studies of rotavirus immunity.     

A simian human immunodeficiency virus with a nonfunctional Vpu (deltavpuSHIV(KU-1bMC33)) isolated from a macaque with neuroAIDS has selected for mutations in env and nef that contributed to its pathogenic phenotype

Previous studies have shown that passage of nonpathogenic SHIV-4 through a series of macaques results in the selection of variants of the virus that are capable of causing rapid subtotal loss of CD4(+) T cells and AIDS within 6-8 months following inoculation into pig-tailed macaques. Using a pathogenic variant of SHIV-4 known as SHIV(KU-1bMC33), we reported that a mutant of this virus with the majority of the vpu deleted was still capable of causing profound CD4(+) T cell loss and neuroAIDS in pig-tailed macaques (McCormick-Davis et al., 2000, Virology 272, 112-116). In this study, we have analyzed the tissue-specific changes in the env and nef in one macaque that developed neuroAIDS (macaque 50 O) and in three macaques that developed only a moderate or no significant loss of CD4(+) T cells and no neurological disease (macaques 50 Y, 20220, 20228) following inoculation with DeltavpuSHIV(KU-1bMC33). Sequence analysis of the gp120 region of env isolated from lymphoid tissues (lymph node and spleen) of macaques 50 Y, 20220, and 20228 revealed no consensus amino acid substitutions. In contrast, analysis of the gp120 sequences isolated from lymphoid and CNS tissues (parietal cortex, basal ganglia, and pons) of macaque 50 O revealed numerous amino acid substitutions. The significance of the amino acid substitutions in gp120 was supported by neutralization assays which showed that the virus isolated from the lymph node of macaque 50 O was neutralization resistant compared to the parental SHIV(KU-1bMC33). Analysis of changes in the nef gene from macaque 50 O revealed in-frame deletions in Nef that ranged from 4 to 13 amino acids in length, whereas the nef genes isolated from the other three macaques revealed no deletions or consensus amino acid substitutions. Inoculation of the virus isolated from the lymph node of the macaque which developed neuroAIDS, SHIV(50OLNV), into four pig-tailed macaques resulted in a severe loss of the circulating CD4(+) T cells within 2 weeks postinoculation, which was maintained for up to 20 weeks postinoculation, confirming that this virus had indeed become more pathogenic in pig-tailed macaques. Taken together, these observations suggest that DeltavpuSHIV(KU-1bMC33) has a low pathogenic phenotype in macaques but that individual pig-tailed macaques can select for additional mutations within the Env and Nef which can compensate for the lack of an intact Vpu and ultimately increase its pathogenicity.     

Evaluation of the Affirm Ambient Temperature Transport System for the detection and identification of Trichomonas vaginalis, Gardnerella vaginalis, and Candida species from vaginal fluid specimens

The objective of this study was to measure the performance of the Affirm Ambient Temperature Transport System (ATTS) over time and to estimate the length of time the system can preserve a vaginal specimen containing the three common organisms causing vaginitis: Trichomonas vaginalis, Candida species, and Gardnerella vaginalis (one of the causative agents of bacterial vaginosis). Women with symptoms of vaginitis presenting to one of three clinical centers were evaluated over a 4- to 8-week period. Four simultaneously obtained swabs were collected and tested by the Affirm VPIII assay at time zero with and without a preservative reagent, at 24 h with reagent, and at either 48 or 72 h with reagent. For each of the three organisms, Trichomonas, Gardnerella, and Candida, positivity at each time point was evaluated and compared to that at reference time zero with and without the ATTS. A total of 940 specimens were obtained from the three clinical sites. Eight hundred three were positive for one or more of the three organisms. Gardnerella had the highest overall positive rate (62%), followed by Candida with 18% and Trichomonas at 9%. The percent sensitivity versus control for Trichomonas ranged from 100% at time zero with and without reagent to 91% by 72 h. Gardnerella and Candida sensitivity remained at 100% for each time period. The Affirm VPIII ATTS system performed within 10% of the control swab (no transport reagent) at all four time points (0, 24, 48, and 72 h) for Trichomonas, Gardnerella, and Candida.