Vaginal and endocervical microorganisms in symptomatic and asymptomatic non-pregnant females: risk factors and rates of occurrence

Transcription

1 ORIGINAL ARTICLE /j x Vaginal and endocervical microorganisms in symptomatic and asymptomatic non-pregnant females: risk factors and rates of occurrence C. Tibaldi 1, N. Cappello 2, M. A. Latino 3, G. Masuelli 1, S. Marini 1 and C. Benedetto 1 1) Department of Obstetrics and Gynaecology, University of Turin. Sant Anna Hospital, 2) Department of Genetics Biology and Biochemistry, Section of Medical Statistics, University of Turin and 3) Department of Clinical Pathology and Microbiology, Sant Anna Hospital, Turin, Italy Abstract Physiological or non-physiological factors may affect the vaginal flora. The occurrence of genital microorganisms in non-pregnant females of all ages was studied, as were the risk factors associated with each microorganism. A retrospective analysis of vaginal and endocervical cultures and wet smears from non-pregnant women, between 1996 to 2005, was performed taking into consideration clinical and socio-demographic characteristics. No microorganisms were observed in 55.7% of the individuals studied and 44.3% had positive cultures. There was no microbiological aetiology in 49% of women with genital symptoms. Poor hygiene, chemical irritants, sexual behaviour, vaginal blood, birth control type, and/or the lack of an oestrogen effect may have caused the symptoms. The highest occurrence of Gram-negative bacteria (p <0.01), mainly Escherichia coli, was observed in prepubescent girls. The highest occurrence of Candida species (p <0.01) was in women of childbearing age, and of Gram-positive bacteria (p <0.01) in menopausal women. Adolescents, particularly asymptomatic girls, carried more frequently Ureaplasma urealyticum and Chlamydia trachomatis (p <0.01). Hormonal contraception and consistent condom use was protective against bacterial vaginosis and U. urealyticum colonization. Users of intrauterine devices had an increased risk of bacterial vaginosis or of contracting U. urealyticum, Mycoplasma hominis and Candida species. Genital complaints were an independent indicator of Candida species, Gram-negative and Gram-positive bacteria, Trichomonas vaginalis and bacterial vaginosis. Chlamydia trachomatis infections were often asymptomatic. It is concluded that the hormonal milieu and non-physiological factors are major determinants of the vaginal flora. If diagnosis of genital infections is based on symptoms alone and not on culture results, it may be erroneous. Sexual abuse should be investigated when a child presents with a sexually transmitted disease. Keywords: Bacterial vaginosis, Candida, hormonal status, sexual behaviour, STD, vaginal flora Original Submission: 24 October 2008; Revised Submission: 14 January 2009; Accepted: 14 January 2009 Editor: G. Pappas Article published online: 24 June 2009 Clin Microbiol Infect 2009; 15: Correspondence and reprint requests: C. Tibaldi, C.so Spezia 60, Turin, Italy ceciliatibaldi@hotmail.com Introduction Genital tract infections are one of the most common reasons why women seek gynaecological advice. In the last few decades an increase in these infections, as in sexually transmitted diseases (STDs), has been observed [1]. A physiological microflora of the genital tract may protect against genital infectious diseases and/or the transmission of STDs. A nonphysiological vaginal microflora is associated with several gynaecological problems such as adverse pregnancy outcomes, postpartum endometritis, post-surgical infections and/or pelvic inflammatory diseases [2 4]. The vaginal flora is highly dynamic with a local microbial system. The equilibrium therein may be disturbed by physiological or non-physiological changes, mediated by the hormonal status, sexual behaviour, vaginal blood, foreign bodies and/or concurrent use of medications. Although some studies have investigated risk factors associated with, and/or the effects of the hormonal milieu on, some specific genital microorganisms, to the best of our knowledge there are no recent studies that have analysed the frequency of a wide spectrum of genital microorganisms in a large female population. The objectives of this study were to evaluate the occurrence of genital microorganisms in a large population of non-pregnant females of all ages and to identify the risk factors associated with each microorganism. Journal Compilation ª2009 European Society of Clinical Microbiology and Infectious Diseases

2 CMI Tibaldi et al. Genital microorganisms in non-pregnant females 671 Methods Study population Non-pregnant females at first visit in the outpatient clinic for Genital Tract Diseases from 1996 to 2005 were enrolled in this study. Tests were made in cases of genital complaints, as part of routine investigation into infertility, before invasive procedures, i.e. insertion of intrauterine devices (IUDs), hysteroscopy, transvaginal sonosalpingography, or when a Pap-test was positive for inflammation. Data collection Data concerning the microbiological tests and the sociodemographic/clinical characteristics of the women for whom cervico-vaginal smears were performed were retrospectively collected from computerized microbiology laboratory archives. The socio-demographic/clinical data included: age, ethnic group, education, marital status, sexual history, type of contraception, obstetrical history, and reason for giving the test. Age was categorized as 13, 14 25, 26 35, 36 45, 46 55, 56 65, 66 years. The first group included children who were presumed not to have had sexual intercourse and the last included women who had not had natural menses for at least 10 years before the tests. A sub-analysis of adolescents (14 19 years), generally considered at high risk of STDs [1], was carried out. The cohort was divided into two categories, according to the presence (symptomatic), or absence (asymptomatic), of clear symptoms of lower genital tract infections (vaginal discharge, vulvar or vaginal complaints). At the time of visit, specific microbiological tests had already been performed in all patients, motivated by the patient s history and physical examination. Our study group included females in three different hormonal states: premenarche (young girls who had not yet had their first menses); childbearing age (non-pregnant women after menarche and before menopause); menopause (absence of natural menses for at least 1 year). Microbiological tests All patients were tested with the same methods: vaginal swabs were taken to evaluate the presence of bacterial vaginosis (BV), Candida species, Trichomonas vaginalis, and/or Gram-positive/negative organisms. Endocervical samples were used to evaluate the presence of Chlamydia trachomatis, Ureaplasma urealyticum and/or Mycoplasma hominis. Vaginal smears, collected from the posterior fornix, were analysed by microscopy at 40 magnification to identify T. vaginalis and Candida spp. Another vaginal swab was placed in transport media and cultured by standard laboratory methods for Candida spp. and Gram-negative/positive bacteria, and then Gram-stained. Candida spp. were isolated by selective culture on CHROMagar (Becton Dickinson) at +37 C for 48 h and identified on the appearance of green colonies. The diagnosis of BV was based on three out of four of Amsel s criteria [5]. Genital mycoplasmas were diagnosed using Mycofast Evolution 3 broth culture (International Microbio D.I.D.). The culture was considered positive at >10 3 unit change colour. Chlamydia trachomatis was detected using Transcription Mediated Amplification (TMA) with a DNA probe dedicated to detect ribosomal RNA gene amplicons of C. trachomatis in endocervical samples (Gen-Probe AMP CT TMA test for Chlamydia, biomérieux) Statistical methods Categorical data were compared using either chi-square or Fisher s exact tests. Univariate analyses were performed separately for each of the risk factor variables, to ascertain the crude odds ratios (OR C ) and the 95% CIs (OR C ), estimated by logistic analysis. A calculation of the adjusted OR (OR A ) and 95% CI (OR A ) was made using multiple logistic regression models, to identify any variables that were independently associated with each single microorganism. Where data were incomplete, the cases were eliminated from the logistic regression models. A p-value of <0.05 was considered statistically significant. Statistical analyses were performed using the SPSS 15.0 software package (SPSS Inc., Chicago, IL, USA). Results A total of females were evaluated; their characteristics are shown in Table 1. A positive test result was observed in (44.3%): in 9463 (78.6%) a single microorganism was isolated and 2577 (21.4%) had polymicrobial colonization. Candida spp. were isolated in 4904/ (18.1%) females, U. urealyticum in 3814/ (16.9%), Gram-positive bacteria in 1778/ (9.1%) and Gram-negative bacteria in 1450/ (7.4%), M. hominis in 385/ (1.7%), T. vaginalis in 382/ (1.6%) and C. trachomatis in 321/ (1.4%); BV occurred in 2135/ (8.9%), The frequency of microorganisms stratified by age is shown in Fig. 1. A total of (54.5%) subjects complained of genital symptoms, among whom a single microorganism was isolated in 5754 (38.8%) and polymicrobial colonization was observed in 1771 (11.9%). No microorganism was observed in 7290 (49.2%) cases. Among (45.5%) asymptomatic subjects

3 672 Clinical Microbiology and Infection, Volume 15 Number 7, July 2009 CMI TABLE 1. Characteristics of the women (n = ) included in the study n % Age (years) (27 172) Country of origin (26 296) Italy Foreigner West Europe East Europe South America North Africa Central West Africa Asia Education (years) (25 780) > Marital status (26 043) Steady partner No steady partner Parity (26 422) Nulliparity Pluriparity Past Preterm labour Past miscarriage Past Abortion/s > Contraception (26 219) None Always condom Sometimes condom/spermicide IUD Oral contraceptives Sexual history Age at first intercourse (years) (24 659) No. of lifetime partners (24 368) > No. of partners in previous 6 months (24 903) > Hormonal state Prepuberty Childbearing age Menopause The denominators used to calculate percentages vary as data were not available for all subjects for all the variables. a single microorganism was isolated in 3666 (30%) cases, polymicrobial colonization was observed in 790 (6.4%) and no microorganism in 7901 (63.9%) cases. Table 2 shows the occurrence of single microorganisms isolated in the symptomatic and asymptomatic females in different periods of their life. All prepubescent girls reported symptoms and had the highest occurrence of Gram-negative bacteria (20.4%, p <0.01). A total of 51.8% of the subjects of child-bearing age were symptomatic, with the highest frequency of Candida spp. (20.5%, p <0.05), and 60.4% of symptomatic menopausal women had the highest occurrence of Gram-positive bacteria (14.8%, p <0.01) and T. vaginalis (2.6%, p <0.01). The sub-analysis of adolescents (data not shown in Table 2) showed that the asymptomatic adolescents had a significantly higher frequency (p <0.01) of U. urealyticum (23.5%) and C. trachomatis (6%) compared with the older fertile women. Univariate and multivariate analyses of some risk factors associated with single microorganisms are shown in Table 3. The multivariate model showed that candidiasis was positively associated (p <0.05) with being years old, being of Central/West African origin, having 9 13 years of education, not having a steady partner, having symptoms and with current IUD use; it was inversely associated (p <0.01) with menopause. Gram-negative bacteria were significantly associated (p <0.01) with having genital symptoms; a negative association (p <0.05) was found with hormonal contraception, age at first intercourse between years, and South- American origin. Gram-positive bacteria were significantly associated (p <0.01) with pluriparity, inconsistent condom or spermicide use and genital symptoms; the association with menopause was of borderline-statistical significance. A negative association (p <0.01) was found with age at first intercourse between years, being of South-American origin and aged between years. Mycoplasma hominis was significantly associated (p <0.05) with not having a steady partner and current use of IUD. BV was significantly associated (p <0.01) with the age range of years, being of East European or Central/West African origin, having 0 8 years of education, not having a steady partner, having had more than one abortion, current use of IUD, age at first intercourse <19 years, more than two lifetime partners, and with genital symptoms. A negative association (p <0.05) was found in individuals without sexual partners in the previous 6 months, or those on oral contraception. Ureaplasma urealyticum was significantly associated (p <0.05) with the age range of years, being of Central/West African origin, having 0 8 and 9 13 years of education, not having a steady partner, having had more than one previous abortion, current use of IUD, inconsistent condom or spermicide use, more than one partner in the previous 6 months, more than two lifetime partners. The positive association with being adolescent was of borderline-statistical significance. A negative association (p <0.05) was found for the use of condom and hormonal contraception. Chlamydia trachomatis was positively associated (p <0.05) with the age ranges of and years, being of South-American or Asian origin, not

4 CMI Tibaldi et al. Genital microorganisms in non-pregnant females % 13 years years years years years years 66 years 20% 15% 10% 5% 0% Candida species Gram negative Gram positive Mycoplasma hominis Bacterial vaginosis Ureaplasma urealiticum Clhamydia Trichomonas trachomatis vaginalis FIG. 1. Frequency of positive cervico-vaginal smears stratified by age: Candida species presented a peak in the years age group. The occurrence of Gram-positive bacteria increased with age, whereas Gram-negative bacteria showed a peak in children ( 13 years). The occurrence of bacterial vaginosis and Ureaplasma urealiticum showed two peaks, between 14 and 25 years of age (9.7% and 22%, respectively) and between 46 and 55 years of age (10.4% and 21%, respectively). Chlamydia trachomatis peaked (2.9%) in patients between 14 and 25 years of age. The frequency of Trichomonas vaginalis increased with age with a peak in the years age group (3.3%). The only cases of T. vaginalis, documented in the 13 years group, were in two sexually abused children. TABLE 2. Positivity for single microorganisms in symptomatic and asymptomatic females according to their hormonal status Hormonal state Prepuberty Childbearing age Menopause Symptomatic Symptomatic Asymptomatic Symptomatic Asymptomatic n % n % n % n % n % Candida spp. 3/ / b 1298/ d 97/ / Gram-positive 2/ / / / c 47/ e Gram-negative 10/ a 417/ / / / M. hominis 10/ / / BV 1/ / / / / U. urealyticum 1110/ / / / C. trachomatis 38/ / / T. vaginalis 86/ / / c 4/ The denominators used to calculate percentages vary because not all women were tested for all microorganisms. a Prepubertal girls had a significantly higher frequency of Gram-negative bacteria compared with childbearing and menopausal women (p <0.01). b Symptomatic childbearing women had a significantly higher frequency of Candida spp. compared with symptomatic prepubertal and menopausal females (p <0.05). c Symptomatic menopausal women had a significantly higher frequency of Gram-positive bacteria and T. vaginalis compared with childbearing and prepubertal females (p <0.01). d Asymptomatic childbearing women had a significantly higher occurrence of Candida spp. compared with asymptomatic menopausal women (p <0.01). e Asymptomatic menopausal women had a significantly higher occurrence of Gram-positive bacteria compared with asymptomatic childbearing women (p <0.05) having a steady partner, having had more than two lifetime partners; it was negatively associated (p <0.01) with symptoms. The positive association with oral contraception or being adolescent were of borderline-statistical significance. Trichomonas vaginalis was positively associated (p <0.05) with the age range of years, being of Central/West African origin, having 0 8 years of education, having 2 10 lifetime partners and having genital symptoms. Discussion This study analyses a wide spectrum of genital microorganisms in a large female population in the age range between 13 and 66 years. Most individuals were Italian, had a lowto medium-level education, a steady partner and a reported low-risk sexual behaviour, and more than half of them used

5 674 Clinical Microbiology and Infection, Volume 15 Number 7, July 2009 CMI TABLE 3. Univariate and multivariate analysis of demographic, behavioural and hormonal factors associated with single microorganisms Candida spp. (n = ) Gram negative bacteria (n = 9050) Gram positive bacteria (n = 9283) Mycoplasma hominis (n = ) Variable (95%CI) for OR OR Age (years) ( ) ( )** ( ) ( )** 1.70 ( )* 0.98 ( ) 1.07 ( ) 1.04 ( ) 1.02 ( ) 1.22 ( ) 1.03 ( ) ( )** 1.57 ( ) ( )** 1.39 ( ) 1.20 ( )* 1.03 ( ) 0.97 ( ) 0.82 ( )* 0.98 ( ) 0.35 ( ) ( )* 1.31 ( ) 1.52 ( )** 1.14 ( ) 1.57 ( )** 1.03 ( ) 6.01 ( )* 2.78 ( ) ( ) 1.15 ( ) 1.62 ( )** 0.85 ( ) 1.82 ( )** 0.93 ( ) 12.2 ( )* ( )* 0.71 ( ) 2.02 ( )** 1.01 ( ) 0 0 Country of origin Italy West Europe 1.02 ( ) 1.00 ( ) 1.06 ( ) 1.08 ( ) 1.03 ( ) 0.96 ( ) East Europe 1.27 ( ) 1.03 ( ) 1.17 ( ) 0.88 ( ) 1.23 ( ) 1.42 ( ) South America 1.03 ( ) 0.90 ( ) 0.44 ( ) 0.20 ( )* 0.40 ( )* 0.30 ( )* North Africa 1.32 ( ) 1.20 ( ) 0.46 ( ) ( ) 0.70 ( ) 5.35 ( ) 11.5 ( )* Central West Africa 2.11 ( )** 1.75 ( )* 0.34 ( ) 0.52 ( ) 1.92 ( )* 1.75 ( ) Asia 1.27 ( ) 1.11 ( ) 0.60 ( ) 0.31 ( ) 0.68 ( ) 0.77 ( ) Education (years) ( )* 1.12 ( ) 1.19 ( ) 0.92 ( ) 1.16 ( ) 0.79 ( ) ( )** 1.15 ( )* 0.97 ( ) 0.87 ( ) 1.11 ( ) 1.05 ( ) > Marital status Steady partner No steady partner 1.38 ( )** 1.35 ( )** 0.87 ( ) 0.99 ( ) 0.99 ( ) 1.07 ( ) 2.89 ( )* 4.93 ( )* Parity Nulliparity Pluriparity 1.32 ( )** 1.41 ( ) 1.64 ( )** 1.34 ( ) 1.75 ( )** 2.03 ( )* 5.26 ( )* 0 Past abortion/s ( ) 0.85 ( )* 0.93 ( ) 0.93 ( ) 1.02 ( ) 0.90 ( ) 2.93 ( ) 1.30 ( ) > ( )** 1.04 ( ) 0.99 ( ) 0.70 ( ) 1.39 ( ) 1.11 ( ) 2.45 ( ) 1.06 ( ) Current contracception None Always condom 1.35 ( )** 1.08 ( ) 1.06 ( ) 1.07 ( ) 0.93 ( ) 0.85 ( ) IUD 2.31 ( )** 1.92 ( )** 1.57 ( )* 1.34 ( ) 0.98 ( ) 0.88 ( ) 7.61 ( )* 5.99 ( )* Oral contraception 1.14 ( )* 0.92 ( ) 0.71 ( )* 0.75 ( )* 0.83 ( ) 0.86 ( ) 1.27 ( ) 1.11 ( ) Sometimes condom/spermicide 1.42 ( )** 1.11 ( ) 0.86 ( ) 0.81 ( ) 1.55 ( )* 1.49 ( )** Sexual history Age at first intercourse (years) ( ) 1.19 ( ) 0.43 ( ) 0.71 ( ) 0.54 ( ) 0.57 ( ) ( )** 0.93 ( ) 0.44 ( )** 0.55 ( ) 0.50 ( )** 0.49 ( )** ( )** 1.04 ( ) 0.59 ( )** 0.50 ( )** 0.55 ( )** 0.55 ( )** ( ) 1.05 ( ) 0.72 ( )* 0.64 ( )** 0.69 ( )* 0.63 ( )** > No. of lifetime partners ( )** 1.07 ( ) 0.74 ( )** 0.85 ( ) 0.77 ( )** 0.96 ( ) 1.26 ( ) 0.92 ( ) ( )** 1.15 ( ) 0.84 ( ) 1.02 ( ) 0.79 ( ) 1.02 ( ) 2.5 ( ) 1.23 ( ) > ( ) 1.19 ( ) 0.38 ( ) 0.58 ( ) 1.23 ( ) 1.51 ( ) 6.22 ( ) 5.00 ( ) No. of partners in the previous 6 months ( ) 0.81 ( ) 1.47 ( )* 1.29 ( ) 1.49 ( )* 1.08 ( ) 2.39 ( ) 1.44 ( ) > ( )* 1.13 ( ) 0.70 ( ) 0.82 ( ) 0.99 ( ) 0.88 ( ) Hormonal status Prepuberty 0.35 ( ) ( )** ( ) 0

6 CMI Tibaldi et al. Genital microorganisms in non-pregnant females 675 TABLE 3. Continued Candida spp. (n = ) Gram negative bacteria (n = 9050) Gram positive bacteria (n = 9283) Mycoplasma hominis (n = ) Variable (95%CI) for OR OR Childbearing age Menopause 0.46 ( )** 0.54 ( )** 1.47 ( )** 1.08 ( ) 1.82 ( )** 1.41 ( ) 3.52 ( ) 0 Sub analysis adolescent Adolescent 1.44 ( )* 1.11 ( ) 0.89 ( ) 1.03 ( ) 1.14 ( ) 0.99 ( ) Childbearing age Symptoms Symptomatic 2.05 ( )** 1.85 ( )** 1.84 ( )** 1.77 ( )** 1.91 ( )** 1.67 ( )** 2.46 ( ) 2.04 ( ) Asymptomatic Bacterial vaginosis (n = ) Ureaplasma urealyticum (n = ) Chlamydia trachomatis (n = ) Trichomonas vaginalis (n = ) Variable (95%CI) Age (years) ( ) ( ) 1.04 ( ) 1.66 ( )** 1.41 ( )* 3.19 ( )** 2.59 ( )* 1.58 ( ) 2.02 ( )* ( )* 2.21 ( )* ( ) 1.04 ( ) 1.08 ( ) 0.89 ( ) ( ) 0.70 ( ) ( )** 1.61 ( )** 1.48 ( )** 1.23 ( ) 2.18 ( ) 3.50 ( )* 3.01 ( )** 1.42 ( ) ( ) 1.02 ( ) 0.87 ( ) 0.40 ( ) 2.27 ( ) ( )** 1.07 ( ) ( )* 0.20 ( )* 0.38 ( ) 0.35 ( ) ( )* 1.03 ( ) Country of origin Italy West Europe 1.04 ( ) 0.94 ( ) 1.29 ( ) 1.21 ( ) 0.48 ( ) 0.46 ( ) 0.41 ( ) 0.50 ( ) East Europe 1.75 ( )* 1.68 ( )* 1.40 ( ) 1.46 ( ) 1.97 ( ) 0.84 ( ) 0 0 South America 1.40 ( ) 1.26 ( ) 1.21 ( ) 1.24 ( ) 3.60 ( )* 4.31 ( )* 0.54 ( ) 0.60 ( ) North Africa 1.41 ( ) 0.91 ( ) 1.59 ( )* 1.42 ( ) 1.38 ( ) ( )* 1.93 ( ) Central West Africa 2.71 ( )** 2.01 ( )* 2.74 ( )** 2.22 ( )* 2.69 ( ) 1.5 ( ) 5.82 ( )** 6.27 ( )* Asia 0.75 ( ) 1.32 ( ) 1.32 ( ) 0.58 ( ) 8.52 ( )** 7.82 ( )* 1.85 ( ) 3.11 ( ) Education (years) ( )** 1.49 ( )** 1.66 ( )** 1.59 ( )** 0.79 ( ) 0.74 ( ) 5.88 ( )** 4.01 ( )* ( ) 1.19 ( ) 1.35 ( )** 1.30 ( )* 1.12 ( ) 0.91 ( ) 1.84 ( ) 1.77 ( ) > Marital status Steady partner No steady partner 1.69 ( )** 1.71 ( )** 1.68 ( )** 1.58 ( )** 2.61 ( )** 2.40 ( )** 1.36 ( ) 1.22 ( ) Parity Nulliparity Pluriparity 1.52 ( )** 0.34 ( ) 1.28 ( )** 1.15 ( ) 1.16 ( ) 3.18 ( ) 3.95 ( )** 0 Past abortion/s ( )** 1.19 ( ) 1.40 ( )** 1.11 ( ) 1.59 ( ) 0.80 ( ) 2.74 ( )** 1.49 ( ) > ( )** 1.46 ( )* 1.75 ( )** 1.32 ( )* 1.73 ( ) 0.90 ( ) 2.42 ( )* 0.68 ( ) Current contracception None Always condom 0.78 ( )* 0.60 ( )** 0.94 ( ) 0.80 ( )* 1.33 ( ) 1.39 ( ) 0.41 ( )* 0.53 ( ) IUD 2.45 ( )** 1.91 ( )** 2.33 ( )** 2.21 ( )** 0.66 ( ) 0.37 ( ) 2.24 ( )* 1.49 ( ) Oral contraception 0.86 ( ) 0.69 ( )** 0.95 ( ) 0.81 ( )* 1.68 ( )* 1.51 ( ) 0.53 ( )* 0.56 ( ) Sometimes condom/spermicide 0.94 ( ) 0.77 ( ) 1.55 ( )** 1.39 ( )** 1.50 ( ) 1.45 ( ) 1.19 ( ) 0.94 ( )

7 676 Clinical Microbiology and Infection, Volume 15 Number 7, July 2009 CMI TABLE 3. Continued Bacterial vaginosis (n = ) Ureaplasma urealyticum (n = ) Chlamydia trachomatis (n = ) Trichomonas vaginalis (n = ) Variable (95%CI) Sexual history Age at first intercourse (years) ( )** 2.05 ( )* 2.11 ( )* 1.10 ( ) 7.46 ( )* 3.81 ( ) 5.93 ( )** ( )** 1.86 ( )** 1.86 ( )** 0.90 ( ) 6.47 ( )* 2.85 ( ) 1.27 ( ) 3.14 ( ) ( )** 1.50 ( )* 1.19 ( ) 0.75 ( )* 3.39 ( ) 1.89 ( ) 0.89 ( ) 0.63 ( ) ( ) 1.11 ( ) 0.92 ( ) 0.69 ( )* 2.55 ( ) 1.6 ( ) 0.82 ( ) 0.68 ( ) > No. of lifetime partners ( )** 1.61 ( )** 1.52 ( )** 1.48 ( )** 2.50 ( )** 1.87 ( )* 1.28 ( ) 2.02 ( )* ( )** 1.90 ( )** 1.72 ( )** 1.70 ( )** 3.71 ( )** 2.84 ( )* 1.94 ( )* 4.22 ( )** > ( )** 2.83 ( )** 1.88 ( )** 1.63 ( )* 5.55 ( )** 4.32 ( )* 2.07 ( ) 2.81 ( ) No. of partners in the previous 6 months ( ) 0.65 ( )* 1.44 ( )* 1.21 ( ) 0.86 ( ) 0.53 ( ) 0.32 ( )* 1.27 ( ) > ( )* 11.8 ( ) 1.95 ( )** 1.47 ( )* 1.98 ( ) 1.03 ( ) 1.84 ( ) 0.72 ( ) Hormonal status Prepuberty 0.36 ( ) Childbearing Menopause 0.81 ( ) 0.85 ( ) 1.02 ( ) 1.11 ( ) 0.46 ( ) ( )** 1.66 ( ) Sub analysis adolescent Adolescent 1.12 ( ) 0.87 ( ) 2.20 ( )** 1.37 ( ) 4.22 ( )** 2.70 ( ) 0.53 ( ) 0.35 ( ) Childdbearing age Symptoms Symptomatic 1.66 ( )** 1.38 ( )** 1.21 ( )** 1.00 ( ) 0.48 ( )** 0.31 ( )** 4.14 ( )** 3.89 ( )** Asymptomatic OR adjusted for all risk factors considered. *p <0.05, **p <0.01

8 CMI Tibaldi et al. Genital microorganisms in non-pregnant females 677 no contraception. We were able to demonstrate a link between Gram-negative bacteria and prepuberty and between Gram-positive bacteria and menopause; Candida spp., BV and U. urealyticum were linked to childbearing age. The predominant detection of Gram-negative bacteria, particularly E. coli, in the prepubescent girls can probably be attributed to poor hygiene and a relatively atrophic vaginal mucosa and/or neutral ph, which provide an excellent environment for growth of enteric bacteria [6,7]. In agreement with other studies [8 10], we found that candidiasis was uncommon in prepuberty, probably because of the specific endocrine milieu. Trichomoniasis was diagnosed only in two prepubescent girls who had been sexually abused. We feel that sexual abuse should always be investigated when a child presents with STDs. There was an overall frequency of Candida spp. of 18.1% at childbearing age; it was the most commonly isolated microorganism, particularly in adolescents, in agreement with findings of other authors [11,12]. Vulvo-vaginal candidiasis is an oestrogen-dependent disorder: oestrogens enhance both Candida adherence to vaginal epithelium and yeast mycelium formation [12]. Ureaplasma urealyticum, the second most frequently detected microorganism, was independently associated with high-risk sexual behaviour. Literature data concerning this association are contradictory [13 16]. The occurrence of BV (8.9%) observed in the present study was lower than that reported in some studies from the USA [17], which may be due to an underrepresentation of Central/West African individuals in the population studied here. The percentage of BV cases increased with age within the childbearing period, as observed by others [18,19], and was independently associated with high-risk sexual behaviour. Although the underlying causes of BV are still to be defined, the relationship between sexual activity and BV has already been demonstrated [20,21]. The presence of BV may also be influenced by the endocrine milieu, as suggested by the 30% reduction in BV among women using oral contraceptives. This effect, observed by other authors [7,22,23], can probably be attributed to the increase in the epithelial cell glycogen content due to the oestrogens, enhancing lactic acid production and, therefore, lowering the vaginal ph [24]. The highest rate of C. trachomatis (3.6%) in women of childbearing age was observed in the subgroup of adolescents and rose to 6% in asymptomatic subjects. The high risk in adolescents may be attributable to sexual behavioural attitudes, anatomical and/or hormonal factors, i.e. an immature cervix, alterations in cervical mucus and/or anovulatory cycles [25]. Gram-positive bacteria were the most frequently isolated microorganisms in the menopause group, probably due to the hypoestrogen-related changes of the vaginal environment (i.e. variations in biofilm formation, thinning of the vaginal epithelium, changes in quantity, viscosity and glycoprotein content of the mucus) and to variations in the immune response [26,27]. Trichomonas vaginalis infection increased with age, but not with menopause, as observed by others [28]. This suggests that it is due to sexual behaviour rather than to hormonal changes. However, in vitro studies [29,30] suggest that the infectivity of T. vaginalis increases at ph and is inhibited by oestrogens. Women over 66 years had a low BV risk, as reported by others [18,26], suggesting that the BV frequency does not increase despite the increase in vaginal ph and despite the fact that lactobacilli are replaced by Gram-positive bacteria and coliforms. It seems plausible that other, not yet defined conditions in the postmenopausal vaginal environment play an important physiopathogenetic role. Age was an independent risk factor: individuals years of age had a significantly higher risk for Candida spp., U. urealyticum, T. vaginalis and/or C. trachomatis than did older women, and those over 66 years showed a significant reduction in BV. The association between the age group of years and C. trachomatis was an unexpected finding and may be related to inaccurate reporting of sexual behaviour. Consistent with literature reports [14,17,28,31,32], race and/or ethnicity were independent risk factors: Central/West African women had an increased risk of BV, U. urealyticum, M. hominis, T. vaginalis and Candida spp, East-European women of BV and Asian and South-American women of C. trachomatis. Low educational levels were independently associated with Candida spp., BV, U. urealyticum and T. vaginalis. Ethnic origin and education were probably associated with contextual characteristics, i.e. poverty, hygiene and alimentary habits which were not recorded in our database. In agreement with those of others, our data suggest that oral contraception has a protective effect against Gramnegative bacteria and BV [7,22,23] and that it does not increase the risk of contracting Candida spp. [33,34]. Hormonal contraception was not found to be a risk factor for C. trachomatis. Literature data concerning this association are contradictory [35 37]. Consistent with data from the literature [12,19,35], there was a significantly higher occurrence of BV, U. urealyticum and Candida spp. in IUD users, as well as a borderline significantly increased risk of C. trachomatis and T. vaginalis, perhaps due to the low occurrence of STDs in the population studied here. Spermicide and/or inconsistent condom use were independent risk factors for Gram-positive bacteria that may be due to changes in the vaginal flora [38]. It was

9 678 Clinical Microbiology and Infection, Volume 15 Number 7, July 2009 CMI observed that the higher the number of lifetime sexual partners, the higher was the risk of acquiring STDs, BV and/or U. urealyticum. Although this study is not without limitations (i.e. it is retrospective, complete data were not available concerning hygiene, alimentary and smoking habits and/or the use of hormonal replacement therapy; some microorganisms were not identified, as selected microbiological tests were performed on the basis of the anamnesis and physical examination), the cohort was very large. The resulting data emphasize that hormonal milieu and behavioural attitudes are the major determinants of vaginal flora. Our data show that the presence of genital complaints are a strong indicator of Candida spp., Gram-negative or Grampositive bacteria, T. vaginalis and BV. Although all prepubescent girls were symptomatic, 67% had no pathogenic vaginal microorganisms and only a single pathogenic microorganism was isolated in the remaining 33%. This result is consistent with those reported by others [8,9] and suggests that most genital symptoms may be due to poor hygiene, chemical irritants and/or the lack of an oestrogen effect. In about half of the symptomatic fertile and menopausal women there was no microbiological aetiology of the symptoms. Therefore, a presumptive diagnosis of genital infections, based on symptoms alone, may be erroneous. Consequently, we are of the opinion that a thorough physical examination and a detailed anamnesis must be done by specially trained health personnel. This should include the recording of all complaints, age, hormonal pattern and sexual behaviour, and the search for an infectious aetiology in order to guide the choice of the correct microbiological tests and of the empiric treatment. Moreover, the data obtained will guide the set-up of preventive/informative strategies to maintain vaginal homeostasis with the aim of reducing the occurrence of genital infections and STDs. Acknowledgments The authors thank B. Wade for her linguistic advice. Transparency Declaration There were no sources of funding for this study. All the authors state that there are no conflicts of interest. The corresponding author (C. Tibald) states that she had full access to all the data in the study and was responsible for the final decision to submit for publication. References 1. Vermillon ST, Homes MM, Soper DE. Adolescents and sexually transmissible diseases. Obstet Gynecol Clin North Am 2000; 27: Goldenberg RL, Andrews WW, Yuan AC et al. Sexually transmitted diseases and adverse outcomes of pregnancy. Clin Perinatol 1997; 24: Penney GC. Preventing infective sequelae of abortion. Hum Reprod 1997; 12 (suppl 11): Sweet RL. Role of bacterial vaginosis in pelvic inflammatory disease. Clin Infect Dis 1995; 20 (suppl 2): S271 S Amsel R, Totten PA, Spigel CA et al. Nonspecific vaginitis diagnostic criteria and microbial and epidemiologic associations. Am J Med 1983; 74: Kokotos F. Vulvovaginitis. Pediatr Rev 2006; 27: Gupta K, Hillier SL, Hooton TM et al. Effects of contraceptive method on the vaginal microbial flora: a prospective evaluation. J Infect Dis 2000; 181: Piippo S, Lenko H, Vuento R. Vulvar symptoms in paediatric and adolescent patients. Acta Paediatr 2000; 89: Jaquiery A, Stylianopoulos A, Hogg S et al. Vulvovaginitis: clinical features, aetiology, and microbiology of the genital tract. Arch Dis Child 1999; 81: Banerjee K, Curtis E, de San Lazzaro C et al. Low prevalence of genital candidiasis in children. Eur J Clin Microbiol Infect Dis 2004; 23: Rein MF. Vulvovaginitis and cervicitis. In: Mandell GL, Bennett JE, Dolin R, eds, Principle and practice of infectious diseases. Philadelphia: Churchill Livingstone, 2000: Ferrer J. Vaginal candidosis: epidemiological and etiological factors. Int J Gynecol Obstet 2000; 71: S21 S Schlicht MJ, Lovrich SD, Sartin JS et al. High prevalence of genital Mycoplasmas among sexually active young adults with urethritis or cervicitis symptoms in la Crosse, Wisconsin. J Clin Microbiol 2004; 42: Newton ER, Piper JM, Shin RN et al. Predictors of the vaginal microflora. Am J Obstet Gynecol 2001; 184: Iwasaka T, Wada T, Kidera Y et al. Hormonal status and Mycoplasma colonization in the female genital tract. Obstet Gynecol 1986; 68: Mardth PA, Westrom L. T-Mycoplasma in the genito-urinary tract of the female. Acta Path Scand 1970; 78: Allsworth JE, Peipert JF. Prevalence of bacterial vaginosis national health and nutrition examination survey data. Obstet Gynecol 2007; 109: Cauci S, Driussi S, De Santo D et al. Prevalence of bacterial vaginosis and vaginal flora changes in peri- and postmenopausal women. J Clin Microbiol 2002; 40: Moi H. Prevalence of bacterial vaginosis and its association with genital infections, inflammation, and contraceptive methods in women attending sexually transmitted disease and primary health clinics. Int J STD AIDS 1990; 1: Nilsson U, Hellberg D, Skoubnikova M et al. Sexual risk factors associated with bacterial vaginosis and Chlamydia trachomatis infection. Sex Transm Dis 1997; 24: Bradshaw CS, Morton AN, Garland SM et al. Higher-risk behavioural practices associated with bacterial vaginosis compared with vaginal candidiasis. Obstet Gynecol 2005; 106: Holzman C, Leventhal J, Qui H et al. Factors linked to bacterial vaginosis in non pregnant women. Am J Public Health 2001; 91: Schwebke JR, Desmond RA, Oh MK. Predictors of bacterial vaginosis in adolescent women who douche. Sex Transm Dis 2004; 31: Skarin A, Sylwan J. Vaginal lactobacilli inhibiting growth of Gardnerella vaginalis, Mobiluncus and other bacterial species cultured from

10 CMI Tibaldi et al. Genital microorganisms in non-pregnant females 679 vaginal content of women with bacterial vaginosis. APMIS 1986; 94: Peipert JF. Genital chlamydial infections. N Engl J Med 2003; 349: Spinillo A, Bernuzzi AM, Cevini C et al. The relationship of bacterial vaginosis, candida and trichomonas infection to symptomatic vaginitis in postmenopausal women attending a vaginitis clinic. Maturitas 1997; 27: Hillier S, Lau RJ. Vaginal microflora in postmenopausal women who have not received estrogen replacement therapy. Clin Infect Dis 1997; 25 (suppl 2): S123 S Kent LH. Epidemiology of vulvovaginitis. Am J Obstet Gynecol 1991; 165: Graves A, Gardner WA. Pathogenicity of Trichomonas vaginalis. Clin Obstet Gynecol 1993; 36: Garber GE, Lemchuk-Favel LT, Bowie WR. Isolation of a celldetaching factor of Trichomonas vaginalis. J Clin Microbiol 1989; 27: Gerbase AC, Rowley JT, Heymann DHL et al. Global prevalence and incidence estimates of selected curable STDs. Sex Transm Inf 1998; 74 (suppl 1): S12 S Center for Disease Control and Prevention. STD surveillance Atlanta, GA: US. Department of Health and Human Sciences, September Nelson AL. The impact of contraceptive methods on the onset of symptomatic vulvovaginal candidiasis within the menstrual cycle. Am J Obstet Gynecol 1997; 176: Spinillo A, Capuzzo E, Nicola S et al. The impact of oral contraception on vulvovaginal candidiasis. Contraception 1995; 51: Beaten JM, Nyange PM, Richardson BA et al. Hormonal contraception and risk of sexually transmitted disease acquisition: results from a prospective study. Am J Obstet Gynecol 2001; 185: Richey C, Macaluso M, Hook E. Determinants of reinfection with Chlamydia trachomatis. Sex Transm Dis 1999; 26: Park B, Stergachis A, Scholes D et al. Contraceptive methods and the risk of Chlamydia trachomatis infection in young women. Am J Epidemiol 1995; 142: Soper DE, Brockwell NJ, Dalton HP. Evaluation of the effects of a female condom on the female lower genital tract. Contraception 1991; 44: