This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Introduction
The vaginal microbiota demonstrates its important role in the reproductive age women’s health or disease. Many studies show that Lactobacillus spp. dominates the vaginal microbiota of healthy women (Ravel et al. 2011; Gajer et al. 2012; Kiecka et al. 2021). Due to Lactobacillus spp. colonization, a pH value up to 4 is maintained, and the binding of pathogenic microorganisms to vaginal epithelial cells is blocked. Bacteriocins, HO, 22and lactic acid produced by lactobacilli can inhibit the growth of different microorganisms (Aroutcheva et al. 2001; Vallor et al. 2001; Karaoğlu et al. 2003; Amabebe and Anumba 2018).
Ureaplasma spp. are frequently isolated from the genital tract of women at childbearing age (Leli et al. 2018). Based on nucleic acid amplification techniques (NAAT) Ureaplasma urealyticum species was split into Ureaplasma parvum and U. urealyticum. Despite the many studies conducted, it has not been resolved whether Ureaplasma spp. are pathogenic or commensal. Some authors describe Ureaplasma spp. as pathogenic in nongonococcal urethritis in men (NGU), pyelonephritis, endometritis (Kundsin et al. 1996; Dewan et al. 1997), and pregnancy complications (Horowitz et al. 1995; Abele-Horn et al. 1997; Harada et al. 2008) while others consider them commensal (Donders et al. 2017). These problems in determining their impact on women’s health may be related to the design of a study and the diagnostic technique used: culture, serology, or polymerase chain reaction (PCR), among others.
The aim of this study was to assess the prevalence of Ureaplasma spp. colonization in a large group of women of childbearing age reporting to a microbiology laboratory for follow-up examinations or by presenting symptoms (itching, burning, discharge, pain, or discomfort). In addition, the objective was to determine the prevalence and co-occurrence of other microorganisms, both common components of the genital tract microbiota and pathogens.
Experimental
Materials and Methods
Study population. This study was a retrospective analysis of the populations of women aged 18 to 49 who had microbiological tests of the genital tract in the Microbiology and Autovaccine Research Center in Cracow, Poland, from April to August 2019, as recommended by a gynecologist. All included patients had to undergo a set of tests such as pH determination, microscopic examination of vaginal smear, microbial culture, and molecular testing for Chlamydia trachomatis, Mycoplasma hominis, U. urealyticum, U. parvum, Mycoplasma genitalium. Moreover, a thorough history of reported symptoms and previous antibiotic treatment were recorded. Women with a history of antibiotic treatment within the past two months were excluded from the study. Information on antibiotic treatment was obtained from the worksheets and referrals for examinations from attending physicians. The following patient data were used for analysis: age, symptoms-reported complaints, pH of vaginal contents, Lactobacillus presence or absence, presence of other bacteria, yeast-like fungi, and Trichomonas vaginalis.
The clinical signs divide the patients into symptomatic and asymptomatic groups. The group of symptomatic women included those who reported at least one symptom in the genital tract: itching, burning, discharge, pain, or discomfort. The asymptomatic group reported no such complaints.
Microbiological methods. The material for microbiological examination consisted of two vaginal swabs (one for Gram-staining microscopy and one for culture) and one cervical swab for molecular testing (NAAT). The vaginal fluid pH was measurand during sample collecting. All tests were performed at the Microbiology and Autovaccination Research Center in Cracow, Poland.
Swabs were taken from the posterior vaginal vault and endocervical canal using a sterile speculum. The vaginal content remaining on the speculum was used to measure the pH value using pH indicator strips with a pH range of 2.0–9.0 (MERCK, Germany). One vaginal swab was used for Gram staining with PREVI® Color Gram stainer (bioMerieux, France). The vaginal Gram stain smears were assessed using the scale according to Kuczyńska modified by Kasprowicz (Kasprowicz and Białecka 2012). The presence of leukocytes, lactobacilli, other bacteria, fungi, and clue cells was assessed based on an abundance scale ranging from 0 to 3 in oil-immersion microscope fields (× 1000). The degrees III, III/IV, and IV with clue cells detected indicated BV, which is comparable to a score from 7 to 10 according to Nugent criteria (Nugent et al. 1991; Kuczyńska 2003; Kasprowicz and Białecka, 2008). The specimen from the first vaginal swab was also inoculated in Trichomedium (GRASO, Poland). After 48 hours of culture, a direct preparation was made and viewed under a microscope (200 × magnification) to identify T. vaginalis.
The second vaginal swab was used to perform cultures on TSA with 5% sheep blood (GRASO, Poland) for the detection of aerobic bacteria, Gardnerella Agar (bio-Merieux, France) for anaerobic bacteria, CANDIselect (BioRad, USA) for yeasts, and Chocolate Agar + P.V.S. + VCAT (BioRad, USA) for Neisseria gonorrhoeae. The media were incubated for 24 h or 48 h at 35°C under appropriate conditions. The cultured microorganisms were isolated and identified using VITEK2 (bioMerieux, France) or MALDI-TOF MS (Brucker, USA).
Material collected from the cervical swabs was preserved on UTM medium (Copan Italia, Italy) and tested for C. trachomatis, M. hominis, U. urealyticum, U. parvum, and M. genitalium. The DNA was isolated with croBEE 201A Nucleic Acid Extraction Kit (Gene-Proof, Czech Republic) with croBEE NA16 Nucleic Acid Extraction System (GeneProof, Czech Republic) according to the manufacturer’s instructions. The NAAT was performed with a commercially available PCR – AmpliSens test, according to the manufacturer’s instructions in a Cobas Z480 real-time thermocycler (ROCHE, Switzerland).
Statistical analysis. A language and environment for statistical computing R were used for statistical analysis (R Core Team 2022). The significance level was assumed at p = 0.05. The Chi-square test of independence, Fisher exact test (for qualitative features) and Mann-Whitney U test (for quantitative features) were used to compare distributions of variables across two independent groups. The odds ratios [ORs] with 95% confidence intervals [CIs] were calculated with Firth’s bias reduction method of logistic regression with the Wald test (Wang 2014, Heinze et al. 2022). The multivariable logistic regression model was built and evaluated to associate various independent factors, such as the presence of different microorganisms or the age of women, with clinical symptoms or elevated pH. The Akaike information criterion was extracted and based on this statistic best-suited (with minimal AIC), the most parsimonious models were selected from a collection of fitted models. Bearing in mind the possible coexistence of some microorganisms, we tested second-order interactions between them. Due to non-significant results, none of the interaction terms was incorporated into the final model.
Results
Study patients’ population. A total of 1,155 women aged from 18 to 49 years were included in the study. Seven hundred fifty-six women in the study group did not report symptoms, while 399 women presented symptoms.
Among 1,155 women, Lactobacillus spp. was found in vaginal swabs of 1,107 (95.8%) (single/few/numerous lactobacilli). The differences in the abundance of Lactobacillus spp. in the patients’ group (n = 1,107) were analyzed; 964 women presented the vaginal smear with few or numerous lactobacilli, and 143 presented only single lactobacilli.
Ureaplasma spp. was detected in 274 women (23.7%). Aerobic bacteria were identified in 153 vaginal swabs (13.2%), and yeast-like fungi in 192 (16.6%). Gardnerella vaginalis was found in 56 specimens (4.8%) from patients with vaginosis. M. hominis (1.8%) and sexually transmitted pathogens C. trachomatis (0.7%) and M. genitalium (0.2%) were found rarely in the study group. T. vaginalis and N. gonorrhoeae, two other etiological agents of STI, were not identified in any specimen in this study. Complete data on genital tract microorganisms’ prevalence are summarized in Table I.
The prevalence of specific microorganisms across groups of women manifesting or not clinical symptoms.
Microorganisms
Microbiological method
Total
Without symptoms (n = 756)
With symptoms (n = 399)
p*
OR*** (95% CI)
p**
n (%)
n (%)
n (%)
Lactobacillus spp.
microscopic
1,107 (95.8)
737 (97.5)
370 (92.7)
< 0.001
0.33(0.18–0.60)
< 0.001
Gardnerella vaginalis
microscopic, culture
56 (4.8)
11 (1.5)
45 (11.3)
< 0.001
8.32(4.30–16.08)
< 0.001
Ureaplasma spp.
PCR
276 (23.9)
108 (14.3)
166 (41.6)
< 0.001
4.26(3.21–5.66)
< 0.001
Ureaplasma parvum
PCR
187 (16.2)
81 (10.7)
106 (26.6)
< 0.001
3.01(2.19–4.14)
< 0.001
Ureaplasma urealyticum
PCR
89 (7.7)
27 (3.6)
62 (15.5)
< 0.001
4.91(3.08–7.83)
< 0.001
Mycoplasma hominis
PCR
21 (1.8)
0 (0.0)
21 (5.3)
< 0.001#
85.94(5.19–1,422.62)
0.002
Mycoplasma genitalium
PCR
2 (0.2)
0 (0.0)
2 (0.5)
0.119#
9.62(0.469.62 –198.69)
0.146
Chlamydia trachomatis
PCR
8 (0.7)
1 (0.1)
7 (1.8)
0.003#
9.62(1.66–55.79)
0.012
Aerobic bacteria (with GBS)
culture
153 (13.2)
95 (12.6)
58 (14.5)
0.396
1.19(0.84–1.68)
0.339
Streptococcus group B (GBS)
culture
88 (7.6)
57 (7.5)
31 (7.8)
0.981
1.04(0.66–1.63)
0.865
Yeast
microscopic, culture
192 (16.6)
100 (13.2)
92 (23.1)
< 0.001
1.97(1.44–2.69)
< 0.001
Neisseria gonorrhoeae
culture
–
–
–
–
–
Trichomonas vaginalis
culture
–
–
–
–
–
* – all p-values denoted by # are from Fisher’s exact test; the remaining comparisons are made with the chi-square test of independence
** – p-values calculated by Wald method using Firth’s logistic regression
*** – OR of prevalence symptoms in women with specific microorganisms detected when compared with those without symptoms
The occurrence of urogenital mycoplasmas (U. parvum, U. urealyticum, and M. hominis) was analyzed regarding multiple colonization with these species and co-infection with C. trachomatis and M. genitalium. The overall incidence of urogenital mycoplasmas in this study group was 24.22%. Most patients (23.02%) presented single mycoplasma species in the genital tract. Both double (1.03%) and triple mycoplasma colonization (0.17%) and co-infections of urogenital mycoplasmas with C. trachomatis and M. genitalium (0.35%) were very rarely observed (Fig. 1).
Fig. 1
The prevalence of mycoplasmas and C. trachomatis according to the multiplicity of colonization in the population studied (n = 1,155).
Urogenital microorganisms and symptoms. Our data show that the presence of different microorganisms in the genital tract is associated with symptoms including itching, burning, discharge, pain, and discomfort. The results of univariable and multivariable logistic regression models are depicted in Table I and II, respectively.
Univariable analysis showed that women in whom G. vaginalis was isolated were more likely to report symptoms (OR = 8.32, 95% CI: 4.30–16.08). Women with identified Ureaplasma spp. were about four times more likely to be symptomatic (OR = 4.26, 95% CI: 3.21–5.66), and a similar trend was found for C. trachomatis, but the group of C. trachomatis positives was small, so estimates were imprecise (OR = 9.62, 95% CI: 1.66–55.79). All 21 women diagnosed with M. hominis were symptomatic in the study group. The data on the diagnosis of M. genitalium was not significant, as it was identified in only two patients. The results also showed a positive correlation between candidiasis and symptoms in the analyzed group of women. In the present study, neither the prevalence of aerobic bacteria, nor Gram-negative and Gram-positive, like Escherichia coli, Klebsiella pneumoniae, Streptococcus spp., Staphylococcus spp., Enterococcus spp., nor the prevalence of GBS was statistically different in women manifesting symptoms compared to the non-symptomatic women. Complete data on the presence of specific microorganisms in the genital tract and their relationship to symptoms are presented in Table I.
Association between the prevalence of microorganisms and clinical symptoms – multivariable analyses. The fully adjusted model and the most parsimonious model.
Microorganisms
Model 1*
Model 2**
OR*** (95% CI)
p**
OR*** (95% CI)
p**
Ureaplasma parvum
3.06(2.17–4.31)
< 0.001
3.07(2.18–4.31)
< 0.001
Ureaplasma urealyticum
5.25(3.20–8.62)
< 0.001
5.20(3.17–8.51)
< 0.001
Mycoplasma hominis
49.30(3.09–787.71)
0.006
48.97(2.94–815.88)
0.007
Chlamydia trachomatis
10.03(1.66–60.42)
0.012
10.04(1.66–60.66)
0.012
Lactobacillus spp.
0.47(0.23–0.98)
0.045
0.48(0.24–0.94)
0.033
Gardnerella vaginalis
3.92(1.89–8.14)
< 0.001
3.87(1.90–7.91)
< 0.001
Yeast
1.73(1.23–2.43)
0.002
1.73(1.23–2.43)
0.002
DF
12
7
AICc
207.5
197.8
ΔAICc
9.7
0.0
* – Model 1 – a fully adjusted model to variables presented in Table II and additionally to Mycoplasma genitalium, Streptococcus group B, aerobic bacteria (with GBS), age, pH > 4.5
** – Model 1 – the most parsimonious model adjusted to variables presented in Table II
*** – OR of prevalence symptoms in women with detected specific microorganisms compared with those without symptoms
Multivariable analysis found the vaginal colonization by G. vaginalis, urogenital Mycoplasma spp., Chlamydia, and yeast-like fungi as independent predictors of vaginitis symptoms, while Lactobacillus spp. was shown as protective against it (Table II).
Ureaplasma parvum and co-infections. Among 1,155 examined women, 274 women had a positive PCR result for Ureaplasma spp., 187 for U. parvum (68%), and 89 cases for U. urealyticum (32%). Both U. urealyticum and U. parvum were diagnosed in two women. Genital tract colonization by U. parvum more than for times increased the chance of bacterial vaginosis with G. vaginalis (OR = 4.30, 95% CI: 2.44–7.49), almost tripled the prevalence of M. hominis infection (1.4% vs. 3.7%), and contributed to the more frequently recorded infection by yeast-like fungi (OR = 1.58, 95% CI: 1.07–2.32). However, it also significantly reduced the chance of U. urealyticum detection (OR = 0.11, 95% CI: 0.03–0.45) (Table III).
Relationship between the presence of U. parvum and other microorganisms.
Microorganisms
Without Ureaplasma parvum (n = 968)
With Ureaplasma parvum (n = 187)
p*
OR*** (95% CI)
p**
n (%)
n (%)
Lactobacillus spp.
930 (9 6.1)
177 (9 4.7)
0.489
0.72(0.35–1.48)
0.372
Garderella vaginalis
32 (3 .3)
24 (1 2.8)
< 0.001
4.30(2.44–7.49)
< 0.001
Chlamydia trachomatis
5 (0.5)
3 (1.6)
0.125#
2.60(0.86–12.81)
0.101
Mycoplasma hominis
14 (1.4)
7 (3.7)
0.064#
2.65(1.05–6.66)
0.0313
Mycoplasma genitalium
2 (0.2)
0 (0.0)
1.000#
0.00(0.05–21.56)
0.534
Ureaplasma urealyticum
87 (9 .0)
2 (1 .1)
< 0.001
0.11(0.03–0.45)
< 0.001
Aerobic bacteria
134 (1 3.8)
19 (1 0.2)
0.214
0.70(0.42–1.17)
0.174
Yeast
150 (1 5.5)
42 (2 2.5)
0.025
1.58(1.07–2.32)
0.019
Streptococcus group B
82 (8 .5)
6 (3 .2)
0.020
0.36(0.14–0.77)
0.017
* – all p-values denoted by # are from Fisher’s exact test; the remaining comparisons are made with the chi-square test of independence
** – p-values calculated by Wald method using Firth’s logistic regression
*** – OR of prevalence of specific microorganisms in women with Ureaplasma parvum compared with those without Ureaplasma parvum
Among the 187 women with positive U. parvum, 81 showed no symptoms, and 106 reported symptoms. The presence of Lactobacillus spp. was analyzed in both groups and was at a similar level (Table I). The prevalence of U. parvum increased the chance of clinical symptoms about three times (OR = 3.07, 95% CI: 2.18–4.31) (Table II).
Ureaplasma urealyticum and co-infections. The colonization of the genital tract by U. urealyticum increased four times the chance of bacterial vaginosis (OR = 4.07, 95% CI: 2.10–7.89). The possibility of M. hominis detection was also significantly increased (OR = 8.00, 95% CI: 3.22–19.86) (Table IV).
Relationship between the presence of U. urealyticum and other microorganisms.
Microorganisms
Without Ureaplasma urealyticum (n = 1,066)
With Ureaplasma urealyticum (n = 89)
p*
OR*** (95% CI)
p**
n (%)
n (%)
Lactobacillus spp.
1,024 (96.1)
83 (9 3.3)
0.319
0.57(0.23–1.37)
0.203
Garderella vaginalis
43 (4 .0)
13 (1 4.6)
< 0.001
4.07(2.10–7.89)
< 0.001
Chlamydia trachomatis
8 (0.8)
0 (0.0)
1.000#
0.00(0.04–12.15)
0.412
Mycoplasma hominis
13 (1.2)
8 (9.0)
< 0.001#
8(3.22–19.86)
< 0.001
Mycoplasma genitalium
1 (0.1)
1 (1.1)
0.148#
5.98(1.24–116.95)
0.025
Ureaplasma parvum
185 (1 7.4)
2 (2 .2)
< 0.001
0.11(0.03–0.45)
< 0.001
Aerobic bacteria
137 (1 2.9)
16 (1 8.0)
0.227
1.48(0.84–2.63)
0.171
Yeast
173 (1 6.2)
19 (2 1.3)
0.272
1.40(0.82–2.39)
0.213
Streptococcus group B
77 (7 .2)
11 (1 2.4)
0.122
1.81(0.88–3.42)
0.083
* – all p-values denoted by # are from Fisher’s exact test; the remaining comparisons are made with the chi-square test of independence
** – p-values calculated by Wald method using Firth’s logistic regression
*** – OR of prevalence specific microorganisms in women with Ureaplasma urealyticum compared with those without Ureaplasma urealyticum
Among 89 women with positive U. urealyticum, 27 were asymptomatic, while 62 women were symptomatic. In all asymptomatic women with U. urealyticum, the presence of Lactobacillus spp. in the vagina was demonstrated (Table I). The prevalence of U. urealyticum increased about five times the chance of clinical symptoms (OR = 5.20, 95% CI: 3.17–8.51) (Table II).
Relationship between the detection of Ureaplasma spp. and pH > 4.5 regarding Gardnerella vaginalis coinfections.
pH ≤ 4.5 n (%)
pH > 4.5 n (%)
OR*** (95% CI)
p**
Ureaplasma spp.
180 (1 5.6%)
94 (8 .1%)
2.1(1.56–2.84)
< 0.005
Ureaplasma Garnerella vaginalis spp. without coinfection
173 (15.7%)
65 (5.9%)
1.60(1.15–2.24)
0.004
** – p-values calculated by Wald method using Firth’s logistic regression
*** – OR of prevalence pH > 4.5 in women with Ureaplasma spp. (without Garderella vaginalis coinfection) compared with those without Ureaplasma spp. infection
Distribution of vaginal pH values in women with and without the presence of specific microorganisms.
Microorganisms
Not prevalent
Prevalent
p*
pH > 4.5 OR (95% CI)**
p**
n
Q2 (Q1-Q3)
n
Q2 (Q1-Q3)
Lactobacillus spp.
48
5.5 (5.0–5.5)
1,107
4.5 (4.0–4.5)
< 0.001
0.01 (0.00–0.05)
< 0.001
Gardnerella vaginalis
1,099
4.5 (4.0–4.5)
56
5.0 (4.9–5.5)
< 0.001
11.18 (6.05–20.65)
< 0.001
Ureaplasma spp.
881
4.5 (4.0–4.5)
274
4.5 (4.5–5.0)
< 0.001
2.09 (1.55–2.82)
< 0.001
Ureaplasma parvum
968
4.5 (4.0–4.5)
187
4.5 (4.5–5.0)
0.001
1.71 (1.21–2.41)
0.002
Ureaplasma urealyticum
1,066
4.5 (4.0–4.5)
89
4.5 (4.5–5.0)
0.004
2.42 (1.55–3.78)
< 0.001
Mycoplasma hominis
1,134
4.5 (4.0–4.5)
21
4.5 (4.5–5.0)
0.008
3.07 (1.31–7.16)
0.010
Aerobic bacteria (with GBS)
1,002
4.5 (4.0–4.5)
153
5.0 (4.5–5.0)
< 0.001
7.44 (5.17–10.69)
< 0.001
Streptococcus group B
1,067
4.5 (4.0–4.5)
88
5.0 (4.5–5.1)
< 0.001
5.89 (3.76–9.24)
< 0.001
Yeast
963
4.5 (4.0–4.5)
192
4.5 (4.5–5.0)
0.003
1.49 (1.06–2.10)
0.023
* – based on Mann-Whitney U test
** – based on logistic regression, crude models, Firth method, OR of prevalence pH > 4.5 in women with specific microorganisms compared with those without their presence Q2 (Q1-Q3) – median (interquartile range)
Association between the prevalence of microorganisms and pH > 4.5 – multivariable analyses.
The fully adjusted model and the most parsimonious model.
Microorganisms
Model 1*
Model 2**
OR (95% CI)
p***
OR (95% CI)
p***
Ureaplasma parvum
1.81 (1.21–2.73)
0.004
1.86 (1.24–2.79)
0.003
Ureaplasma urealyticum
2.23 (1.31–3.78)
0.003
2.32 (1.37–3.92)
0.002
Lactobacillus spp.
0.03 (0.01–0.11)
< 0.001
0.03 (0.01–0.11)
< 0.001
Gardnerella vaginalis
5.91 (2.91–12.00)
< 0.001
6.54 (3.25–13.18)
< 0.001
Aerobic bacteria (with GBS)
7.31 (4.14–12.91)
< 0.001
6.85 (4.61–10.18)
< 0.001
DF
11
5
AICc
287.7
275.0
ΔAICc
12.7
0.0
* – Model 1 – fully adjusted model variables presented in the Table and additionally to yeast, Streptococcus group B, Mycoplasma hominis, Mycoplasma genitalium, Chlamydia trachomatis, and age
** – Model 2 – the most parsimonious model, adjusted to variables presented in the Table; OR of prevalence pH > 4.5
*** – p-values calculated by the Wald method using Firth’s logistic regression
Ureaplasma spp. and the pH value. Among 1,155 vaginal smears, 76.6% of samples showed the correct vaginal pH range from 3.5 to 4.5. In 23.3% of samples, pH was increased > 4.5. The statistical analysis indicated a strong association between Ureaplasma spp. presence in the genital tract with higher pH (Fig. 2, Table VI).
Fig. 2
Distribution of vaginal pH values in women with and without the presence of specific microorganisms. Dots denote extreme values; vertical segments are placed at median values, and boxes are located at quartile values. All differences in pH levels between groups were statistically significant.
According to our data, a higher pH value (pH > 4.5) increases the chance of Ureaplasma spp. detection regardless G. vaginalis infection (Table V).
Discussion
This presented retrospective study showed a set of results from women in reproductive age diagnosed at the Microbiology and Autovaccine Research Center in Cracow, Poland. Gynecologists referred patients participating in the study to the laboratory for preventive or diagnostic examinations in case of suspected infection in the genital tract. The relationship between the presence of specific microorganisms and the manifestation of various symptoms was investigated. The significant positive associations between vaginal infection of G. vaginalis or Candida spp. with symptoms such as itching, burning, discomfort, discharge, and pain were detected. The results connecting BV or candidiasis with typical vaginitis or cervicitis symptoms have also been described by Leli et al. (2018). This study also found that in women with symptoms, C. trachomatis genital tract infections were over nine times more often identified than in the group without symptoms. It may be related to the previously studied occurrence of the pelvic inflammatory disease reported after C. trachomatis infection (O’Connell and Ferone 2016). Statistically significant differences in our study were observed in colonization by U. urealyticum of symptomatic or asymptomatic women as others reported (Hunjak et al. 2014; Marovt et al. 2015). Similarly, our results conformed the observations by De Francesco et al. (2013) on a relationship between symptom onset and infection by Ureaplasma spp.
Among 1,155 women in our study, 24,2% were colonized by single or multiple species of genital mycoplasmas. Most patients (23.02%) presented single mycoplasma species versus infrequent double and triple colonization, which was detected only in 1.2% of women. The other observational study conducted on a group of 1,761 women showed a much higher general prevalence of mycoplasmas (56.4%). In 39.4% of patients, single species colonization was detected (Leli el al. 2018). Contrary to our results, no association between Ureaplasma spp. and symptoms of vaginitis or cervicitis was reported in the study mentioned above.
The study shows that women with Lactobacillus spp. in the vagina reported symptoms less frequently, confirming the long-known protective role of Lactobacillus spp. (Boskey et al. 2001; Ravel et al. 2011; Witkin and Linhares 2017). No relationship was shown between Lactobacillus spp. and infection by Ureaplasma spp. Lactobacillus spp. was presented in women without Ureaplasma spp. and in women with Ureaplasma spp. symptomatic and asymptomatic. There was also no relationship between Ureaplasma spp. and C. trachomatis infection, as identified by Yamazaki et al. (2012) with only a statistically insignificant trend. A similar trend was shown by Leli et al. (2013). Moreover, U. urealyticum colonization in symptomatic women increases the prevalence of M. hominis (Leli et al. 2013).
Several observational studies have shown that BV and undetectable vaginal lactobacilli are important risk factors for STIs, such as HIV (Martin et al. 1999) and HPV infections (Watts et al. 2005; Biernat-Sudolska et al. 2011). BV microbiota assessed by Gram staining is associated with a significantly increased risk for acquisition of T. vaginalis, N. gonorrhoeae, and C. trachomatis infection (Schwebke and Desmond 2007; Brotman et al. 2010). Our results show that the colonization of the genital tract with U. urealyticum and U. parvum increased four times the chance of bacterial vaginosis. These results are consistent with reports on the epidemiology of U. urealyticum and M. hominis infections associated with the presence of BV and BV markers in pregnant women (Donders et al. 2017).
Excessive growth of G. vaginalis with a reduced number of Lactobacillus spp. leads to an increase of vaginal pH (Mączyńska 2008; Kasprowicz and Białecka 2012). The correlation between increased vaginal pH and BV is well established, making pH testing one of the predictors of BV according to Amsel’s criteria (Amsel et al. 1983). In our study, the interesting association between colonization by Ureaplasma spp. and higher pH was detected regardless of co-infection with G. vaginalis indicating bacterial vaginosis. During Ureaplasma spp. colonization, the increase of vaginal pH may be partially mediated by urea hydrolysis into ammonia. Also, M. hominis produces ammonia from arginine, thus alkalizing the environment. Alkalization of the vaginal environment by mycoplasmas is opposite to the acidifying effect of Lactobacillus. Since our retrospective study lacks data on urogenital mycoplasmas and Lactobacillus load, quantitatively comparing their presence with a pH value is impossible. In addition to the microbiota, the vaginal pH can also be affected by hormonal status, sexual, and hygiene behavior. Our results show an interesting relationship between genital mycoplasmas, bacterial vaginosis, and symptoms of genital tract infections that needs further confirmation.
Our many years of experience in microbial diagnostics were starting points for retrospective analysis of microorganisms’ prevalence, including urogenital mycoplasmas, in the south Poland population of women. The paper is retrospective, based on patients’ charts who underwent routine microbiological diagnosis of the genital tract and answered questions regarding symptoms and antibiotic treatment and medical referrals. The paper attempts to use a large pool of data on the microbiota of the genital tract in a population of women in the south of Poland. Information on the prevalence of specific etiological factors of reproductive tract infections could be valuable for clinics. As a part of routine work, our diagnostic laboratory of the CBMiA carries out diagnostics of the reproductive tract, including microscopic technique, microbial culture technique, and molecular methods (NAAT).
Infections of C. trachomatis, N. gonorrhoeae, and M. genitalium developing in the cervix cause many complications. Colonization of lower genital tract with mycoplasmas can lead to asymptomatic infections of the woman’s upper reproductive system (Kasprzykowska et al. 2014). Often asymptomatic or sparsely symptomatic cervical infections cause complications such as pelvic inflammatory disease (PID), fallopian tube inflammation, fallopian tube obstruction, and ectopic pregnancies. Also, the incidences of sexually transmitted diseases differ regionally; therefore, the knowledge of the occurrence of STI pathogens is valuable. In Poland, women of childbearing age who are sexually active are often referred by gynecologists for prophylactic testing to diagnose infections with C. trachomatis, N. gonorrhoeae, T. vaginalis, but also Ureaplasma spp., and Mycoplasma spp. However, women reporting complaints in the genitourinary tract such as increased vaginal discharge, soreness on urination, soreness during intercourse, and lower abdominal pain are often treated empirically based on the reported symptoms.
The strength of our study is the numerous patient group and the broad picture of microorganisms that have been investigated. On the other hand, the impact of microbiota on women’s health and often co-infections certainly cannot be solved by methods used in routine microbiological diagnostics. The diversity of the vaginal microbiota requires further studies with high-throughput sequencing. However, classic diagnostic microbiology can still broaden the knowledge of the microorganism prevalence in the genital tract and support women’s reproductive health.
In addition, the group of women with symptoms was smaller than the asymptomatic group. It is mainly because of the retrospective character of the analysis and the chosen exclusions parameters. Some women who reported symptoms were not tested with all three microbiological methods; microscopy, culture, and NAAT, and, therefore, were excluded from the analysis. In some cases of STI risk, only NAAT was proceed according to the physician’s prescription. Also, when candidiasis or BV were suspected, only microscopic evaluation of vaginal smear and culture were conducted along with the gynecologist’s recommendation. In our opinion, the picture of coinfections needs more attention from physicians, diagnosticians, and scientists. Future studies should include more numerous groups of women with particular attention to symptoms. The symptoms assessment is an important factor in the analysis. It may result in bias since not every woman experiences the symptoms to the same degree, and many factors can influence this assessment.
In conclusion, a relationship between colonization of the female genital tract by Ureaplasma spp. and coinfection by G. vaginalis were observed in our study group. It was shown that the presence of Lactobacillus spp. significantly reduces the risk of symptoms in women. It was also assessed that BV with G. vaginalis isolation was more often detected in the group of symptomatic women. However, with a view to studies from other groups, further analyses of coinfections and potential associations with symptoms occurring in genital tract infections are needed.
