The prevalence and antimicrobial resistance of Haemophilus inuenzae in prepubertal girls with vulvovaginitis

Background: To determine the prevalence of Haemophilus inuenzae vulvovaginitis in prepubertal girls and the antimicrobial resistance of H.inuenzae strains isolated from vulval specimens. Methods: Isolates of H.inuenzae from vulval swabs of prepubertal girls with vulvovaginitis received at The Children's Hospital, Zhejiang University School of Medicine during 2016-2019 were studied. Vulval specimens were inoculated on Haemophilus selective chocolate agar. Antimicrobial susceptibility tests were performed using the disk diffusion method. A cenase disk was used to detect β-lactamase. Results: A total of 4142 vulval specimens were received during the 4 years, 649 isolates of H. inuenzae were isolated from 642 girls aged 6 months to 13 years, with a median of 5y. There were peaks of isolates from April to July seen in the vulval isolates. In total, the ampicillin resistance rate was 39.1% (250/640); 33.2% strains (211/636) were for β-lactamase-positive isolates, 6.6% strains (42/635) were β-lactamase-negative and ampicillin-resistant (BLNAR) isolates. The resistance rates of H. inuenzae isolates to amoxycillin-clavulanic acid, ampicillin-sulbactam, cefuroxime, ceftriaxone, cefotaxime, meropenem, levooxacin, sulfamethoxazole-trimethoprim, azithromycin, and chloramphenicol were 26.4%, 21.8%, 24.8%, 1.7%, 1.0%, 0.2%, 0%, 47.7%, 10.2%, and 1.1%, respectively. MDR was present in 41 (6.4%) of the 642 H. inuenzae isolates, with the most prevalent MDR phenotype of ampicillin-sulfamethoxazole-trimethoprim-azithromycin


Background
Vulvovaginitis is a common problem in prepubertal girls, which is most commonly attributed to poor hygiene or nonspeci c irritants, but can occur as a part of bacterial infection [1]. Several studies have reported that Streptococcus pyogenes and Haemophilus in uenzae (H. in uenzae) are the most common bacterial causes of juvenile vulvovaginitis [1][2][3][4]. The association between H.in uenzae and prepubertal vulvovaginitis was rst highlighted by MacFarlane in 1987 [5]. But Cox RA's study in 1997 showed that H. in uenzaewas an underrated cause of vulvovaginitis in young girls [4]. However, until now, few studies have comprehensively explored the prevalence of H. in uenzae vulvovaginitis in prepubertal girls. Furthermore, published studies that included a large sample size are also scarce. Therefore, we described a four-year study, undertaken in the laboratory of a tertiary university hospital, to determine the prevalence of H. in uenzae vulvovaginitis in prepubertal girls and the antimicrobial resistance of H.in uenzae strains isolated from vulval specimens.

Strain Collection and Identi cation
This was a retrospective analysis of data from prepubertal girls who presented to the outpatient clinic of pediatric and adolescent gynecology at The Children's Hospital, Zhejiang University School of Medicine between January 2016 and December 2019. Two vulval swabs were taken from each girl, one for microscopic examination and the other for cultivation. The nding of a large number of polymorphonuclear leukocytes on Gram-stained smears indicated the presence of in ammatory reaction.
The specimens were inoculated directly onto Columbia blood agar, Chocolate agar (Haemophilus and Gonorrhoeae selective chocolate agar), and Sabouraud's agar plates in air or 5% CO 2

Statistical Analysis
Antibiotic resistant rates were analyzed with WHONET 5.6. Comparisons of antibiotic resistant rate between groups were performed with the X 2 test. Age data, which were non-normally distributed, were described as medians (IQR). P values < 0.05 were considered statistically signi cant.  Table 1 shows the yearly totals with an increase in numbers received but a decrease in proportion of positives for H. in uenzae. The monthly totals for positive isolates of H. in uenzae show some peaks and troughs throughout the year (Fig. 1). The numbers of isolates per month ranged from 2 in October 2017 to a maximum of 27 in July 2019. There were peaks of isolates from April to July seen in the vulval isolates. The age range of children carrying the H. in uenzae isolates was from 0.5y to 13y but 477 (75%) were between the ages of 3 and 7 years, with a median of 5y (IQR: 3) (Fig. 2).  Table 2). In different years, the resistance rates of H. in uenzae strains to cefuroxime and azithromycin had a signi cantly statistical difference (P < 0.05; Table 3). β-lactamasepositive H. in uenzae strains had signi cantly higher resistance rates to ampicillin, amoxycillin-clavulanic acid, cefuroxime, sulfamethoxazole-trimethoprim, azithromycin, and chloramphenicol than β-lactamasenegative strains did (P < 0.01; Table 4). BLNAR H. in uenzae strains were all resistant to amoxycillinclavulanic acid, ampicillin-sulbactam and cefuroxime while all susceptible to levo oxacin, azithromycin and chloramphenicol (Table 5).

MDR Pattern
MDR was present in 41 (6.4%) of the 642 H. in uenzae isolates. The most prevalent resistance phenotype was ampicillin-sulfamethoxazole-trimethoprim-azithromycin resistance, which was detected in 16 isolates, representing 39% of MDR strains (Table 6).

Discussion
Vulvovaginitis in prepubertal children is a common complained problem in clinical practice. The anatomy of vulva and the thin uncorni ed vulval epithelium at prepubertal age makes it susceptible to infection [1].
Because few hospitals provide specialist paediatric gynaecological outpatient services, children with vulvovaginitis are managed mainly in primary care [3]. Therefore, there were few studies comprehensively explored the prevalence of H. in uenzae vulvovaginitis in prepubertal girls. This study was performed at a tertiary university hospital, which provides specialist paediatric gynaecological outpatient services.
Vulvovaginitis caused by upper respiratory ora is generally considered to be the most common gynecological problem in prepubertal girls. A multicenter study showed that paediatric in ammatory vulvovaginitis is mainly caused by pathogens of the upper respiratory tract and the most common risk factor for this infection is to have suffered an upper respiratory tract infection in the previous month [6]. A study of case report provided the direct evidence of the nose-hand-vagina method of transmission [7]. It has been assumed that respiratory bacteria were transmitted to the perineal area via the hands [8]. Therefore, advice on hygiene and behavior may be an important strategy to prevent vulvovaginitis in prepubertal girls.
Several previous studies have indicated that vulvovaginitis in girls is mostly caused by the bacteria from the upper respiratory tract, S. pyogenes and H. in uenzae [1]. In a large study from Liverpool, H. in uenzae was a more common cause of this complaint than β haemolytic streptococci [9]. However, H. in uenzae is fastidious in its growth requirements and laboratories may not isolate it unless they include appropriate culture medium for genital swabs received for young girls [10]. In the present study, all the specimens were inoculated onto Haemophilus selective chocolate agar for the isolation of H. in uenzae. Published studies have described a large number of Gram-negative and Gram-positive bacteria as the possible causes of vulvovaginitis in girls. However, the pure or predominant growth of a possible pathogenic microorganism, associated with the signs of in ammation, is certainly of diagnostic relevance [1]. In this study, the nding of a large number of polymorphonuclear leukocytes on Gram-stained smears was used to indicate the presence of in ammatory reaction, which ensured that the H. in uenzae isolated from vulval swabs was a possible pathogenic microorganism. There were 649 (15.7%) of 4142 vulval swabs from children yielded H. in uenzae in this study, which was in agreement with the opinions described above that vulvovaginitis in girls is mostly caused by the bacteria from the upper respiratory tract, S. pyogenes and H. in uenzae, suggesting that H. in uenzae was a common pathogen of vulvovaginal infection in prepubertal girls in Zhejiang, China. There were peaks of isolates from April to July seen in the vulval isolates, which was consistent with that from the respiratory tract specimens, suggesting that vaginal H. in uenzae strains might from the respiratory tract [6,11]. The age range of children carrying the H. in uenzae isolates was from 0.5y to 13y but 477 (75%) were between the ages of 3 and 7 years, which were in agreement with the results of previous studies [11,12].
Ampicillin became the drug of choice for the treatment of H. in uenzae infections since the 1970s [13]. In recent years, with the extensive use of antibiotics, the drug resistance of H. in uenzae strains to ampicillin has gradually increased. The ampicillin resistance rate of H. in uenzae strains in China was increased from12% in 2000-2002 [14] to 58.1% in 2016 [15]. In this study, the ampicillin resistance rate was 39.1%, which was higher than that in genital strains (26.4%) but lower than that in respiratory strains (58.4%)in 2015 reported by our study team [15]. In different years, the ampicillin resistance of H. in uenzae strains isolated from vulval specimens had gradually increased, from 33.1% in 2016 to 42.8% in 2018.Therefore, it's worth noting of ampicillin resistance of H. in uenzae in clinical management.
However, the production of β-lactamase is still the main mechanism of ampicillin resistance of H. in uenzae strains in this study, which was consistent with the results in other reports [12,16,17], but different from the mechanism in Japan (BLNAR accounted for > 50% after 2014) [18], while BLNAR accounted for only 6.6% in our study, suggesting great differences in antibiotic resistance and drugresistant mechanisms of H. in uenzae strains around the world. There have been studies comparing the H. in uenzae resistance pro les between respiratory tract and urinary tract [19], respiratory isolates and vaginal isolates [11], the resistance pro les of H. in uenzae vary greatly depending on the infection site, which indicated that the optimal antibiotic treatment for H. in uenzae might vary depending on the region and the site of infection. For BLNAR strains, these drugs should be avoided to use with the reason that BLNAR H. in uenzae strains were all resistant to amoxycillin-clavulanic acid, ampicillin-sulbactam and cefuroxime. The resistance rates of the H. in uenzae isolates to amoxycillin-clavulanic acid, and ampicillin-sulbactam were 26.4%, 21.8% in this study, which might be attributed to BLNAR strains and βlactamase-producing clavulanic acid/amoxicillin-resistant (BLPACR) strains of H. in uenzae. The mechanisms of BLPACR strains might be β-lactamase and PBP amino acid substitutions [20].
Generally, H. in uenzae strains are highly susceptible to third-generation cephalosporins. The nonsusceptibility rate of H. in uenzae to third-generation cephalosporins was under 2% in the present study, which was much lower than that in Iran (33.1%) [21] and Japan (49.4%) [22], but similar to the rate of genital strains in china in 2015(5.5%), a reason for the differences might be explained by the different sites of infection. Typically, H. in uenzae is sensitive to carbapenem; however, carbapenem-nonsusceptible H. in uenzae has been reported previously [23]. In this study, we found one H. in uenzae strain non-susceptible to meropenem; its mechanism is worthy to research in further study.
A high prevalence of sulfamethoxazole-trimethoprim resistance (47.7%) among H. in uenzae isolates was found in this study. However, no signi cant difference was found between the current results (47.7%, 306/641) in 2016-2019 and the previous results in 2015 (51.8%, 57/110) [11], which might be because of fewer applications of sulfamethoxazole-trimethoprim. There were 10.2% of the H. in uenzae isolates resistant to azithromycin in this study, and a signi cantly increased resistance was seen between 2018 and 2019, which might because the extensive use of azithromycin in respiratory infections in China. H. in uenzae strains were all sensitive to levo oxacin in the study; and 1.1% of H. in uenzae strains were resistant to chloramphenicol, which might because of that these antibiotics are not used in children in China. However, o oxacin or levo oxacin gel is still widely used as topical antibiotics in treating local infections in China, including prepubertal vulvovaginitis. MDR was present in 41 (6.4%) of the 642 H. in uenzae isolates. The most prevalent resistance phenotype was ampicillin-sulfamethoxazoletrimethoprim-azithromycin resistance, which was in accordance with the previous ndings [15].

Conclusions
To our knowledge, this study currently represents the largest population-based study of H. in uenzae vulvovaginitis in prepubertal girls in China. H. in uenzae is a common cause of vulvovaginitis in prepubertal girls in Zhejiang, China. Laboratories should ensure that they include media appropriate for Distribution of ages in children carrying the H. in uenzae strains isolated from vulval specimens, 2016-