Arch Gynecol Obstet (2010) 282:43–47 DOI 10.

1007/s00404-009-1241-7

G E N E RA L G Y N E CO L O G Y

Comparison of local metronidazole and a local antiseptic in the treatment of bacterial vaginosis
Aleksandra Novakov Mikic · Dragan Budakov

Received: 17 May 2009 / Accepted: 25 September 2009 / Published online: 7 October 2009 © Springer-Verlag 2009

Abstract Objective Bacterial vaginosis (BV) is characterized by a mixed Xora of pathogenic anaerobic bacteria and associated with risks of pathologic conditions. In the present study, therapy with a local antiseptic spray (octenidine hydrochloride/phenoxyethanol, OHP) for 7 or 14 days is compared against the standard local therapy of BV (metronidazole) in a Serbian patient population. Methods As much as 450 women were treated in groups with either 7 days metronidazole vaginal tablets, 7 days OHP, or 14 days OHP. Control smears were taken after each treatment period. Results In total, 63.2% of the women were without indications of BV after therapy (metronidazole: 61.0%, OHP 7 days: 57.6%, and OHP 14 days: 71.0%). SigniWcantly fewer women were aVected from infections after treatment with 14 days OHP compared to OHP for 7 days. Conclusions Octenidine hydrochloride/phenoxyethanol spray was as eVective as the standard therapy with metronidazole. Patients stated that OHP was more comfortable, easier to apply, and side eVects were lesser. Keywords Bacterial vaginosis · Metronidazole · Octenidine hydrochloride, Phenoxyethanol

Introduction Bacterial vaginosis (BV) is the most common microbiological disorder of the vaginal milieu in women during the reproductive period. The prevalence varies from 5% to 50% [1, 2]; whereas the disease is asymptomatic in approximately half of the patients, BV is an inXammation caused by a combination of several anaerobic bacteria, especially by Gardnerella vaginalis. The disease is associated with an increased risk of several pathologic conditions, including postoperative infection following hysterectomy [3], postabortion pelvic inXammatory disease [4], increased risk for spontaneous abortion, preterm delivery, preterm rupture of membranes, and postpartal endometritis [5]. The risk of plasma cell endometritis in women suVering from BV has been reported to be 15 times higher than the risk for women without BV [6]. Furthermore, it is discussed that BV is connected with the development of cervical intraepithelial neoplasia [7]. By deWnition, BV is diagnosed when at least three out of four criteria resulting from the gynecological examination are tested positive: homogeneous vaginal discharge, the presence of clue cells (>20%), amine odor (Wshy) by adding potassium hydroxide solution to vaginal secretion, a vaginal pH-value above 4.5, and the absence of the normal vaginal lactobacilli [8, 9]. The disease is treated like an anaerobic inXammation. Generally, for systemic and local antibiotic therapy, metronidazole or clindamycin are recommended [10]. Another treatment possibility is the local application of octenidine hydrochloride/phenoxyethanol (octenisept®) as spray [11]. This kind of alternative treatment has not been suYciently supported by clinical trials. In the present study performed within a Serbian patient population, the eVectiveness of a therapy using octenidine

A. Novakov Mikic (&) · D. Budakov Department of Obstetrics and Gynaecology, Clinical Centre of Vojvodina, Branimira Cosica 37, 21000 Novi Sad, Serbia e-mail: aleksandranovakov@yahoo.com

123

Results Out of the 450 patients initially included in the present study. groups 2 + 3: 29. the patients were randomly assigned into three equal groups (Fig. presence of clue cells (more than 20%).3 years (range 18–50 years). All women involved in this research had a regular menstrual cycle. vaginal pH-value above 4. of 139 patients treated with octenidine hydrochloride/phenoxyethanol for 7 days (33.7%.9%. and absence of the normal vaginal lactobacilli. and 138 patients treated for 14 days (33. The aim was to Wnd out the most eVective therapy for the improvement or eradication of BV. For the diagnosis of BV all patients were tested positive in at least three out of the following criteria: homogeneous vaginal discharge. and in group 3 the patients (n = 150) were treated with octenidine hydrochloride/phenoxyethanol spray for 14 days. group 3) have been sampled. The study was performed in accordance with local laws.0 (© SPSS Inc. 450 women with clinically ensured diagnosis of BV were included. in group 2 the patients (n = 150) were treated with octenidine hydrochloride/phenoxyethanol spray over a time period of 7 days. Control smears of 136 patients treated with metronidazole (32. patient characteristics regarding demographic data and gynecological anamnesis (gravidity and parity) were recorded. Statistically signiWcant diVerences in the mean age existed neither between the two therapy groups (group 1 vs. hence. and was comparable in all three patient groups (see Table 1).8%) could be analyzed in the followup visit. Serbia. 413 women (91. group 2). the women were examined gynecologically. Fig. by means of an applicator. group 1). The statistical analyses performed in this study were calculated by using the software SPSS Statistics 13.4%. University of Novi Sad.). Patients and methods The study was a prospective observational study conducted in the ambulant gynecological setting between May 2007 and January 2008.44 Arch Gynecol Obstet (2010) 282:43–47 hydrochloride/phenoxyethanol over a period of 7 and 14 days has been compared against the standard protocol for therapy of BV using metronidazole for 7 days. In addition. Initially. spraying 10 shots of the solution into the vagina during the entire 7 or 14 days of therapy.5. and follow-up characteristics 123 . Table 1 summarizes the characteristics of all patients with follow-up visit. For analyses. respectively. The women in groups 2 and 3 treated with octenidine hydrochloride/phenoxyethanol applied the solution according to the product information: every evening. they were asked verbally by their attending physicians to state their satisfaction with the applied therapy. and amine odor (Wshy) when potassium hydroxide solution was added to the vaginal secretion. 37 patients were lost to follow-up. 1): in group 1 the patients (n = 150) were treated with metronidazole vaginal tablets (1 £ 500 mg daily) for 7 days.4 vs. and control smears were taken for further bacteriologic analysis to check the degree of vaginal discharge. In a follow-up visit after 7 or 14 days. 1 Study overview of patients included. therapy used. and approval was obtained from the ethics committee of the Faculty of Medicine. Additionally. no further data were available. The average age of the patients was 29.

and with Wve pregnancies in 0. However. the majority of women were nulligravida (39.734). 12 women (2. in the statistical analysis between the two octenidine therapy groups (7 vs.4%) 277 65 (47. and there were less women with one pregnancy (30.5%. p < 0.0%) 32 (23. Women with three pregnancies were found in 6.001.1%) two pregnancies.6%) 0 (0.8%) still had bacterial vaginosis (BV) and 261 women (63. 14 patients had a concomitant infection.0% of the women in the metronidazole group had no indications of a bacterial vaginosis (negative control smears) after therapy.5%) 22 (16. and 10 from candidosis.0%) 109 (39. p = 0.5% of all examined women (n = 35.74 0–1 12 (8.2%.3%) three pregnancies.0%) 1 (0. 57. p = 0. No signiWcant statistical diVerence was observed in the number of deliveries with regard to the two patient groups ( 2-test.8%) 9 (6.0% vs.4) 45 Table 2 Comparison of gravidity and parity in diVerent therapy groups Pregnancies/ Gravidity (p = n. no statistically signiWcant diVerence was observed in the number of pregnancies ( 2-test. 16.904) as well as of the number of deliveries (Kruskal–Wallis test. t-test. Additionally. In total.s. signiWcantly less women have been aVected from bacterial vaginosis 123 .866). the results of the control smear did not diVer statistically by comparison of the metronidazole group 1 versus group 2 ( 2-test. respectively.8) 0. p = 0.0%) 136 10 (3. Not signiWcant Concomitant infection 29.4 18–50 29.472). 2). After the corresponding therapy for 7 or 14 days.9%) with one delivery.6%.8% vs.8%) 131 (47. 2.8% vs. 30 women (7. and in group 2 with an octenidine hydrochloride/phenoxyethanol treatment for 1 week.2 years.04 0–5 1.8%) 136 21 (7.s. In the second group.0%) 0 (0. 32. By the comparison of the three patient groups no statistical diVerence was observed regarding the presence of concomitant infections ( 2-test.643) as well as versus group 3 ( 2-test. control smears were taken from all 413 women in the follow-up visits and analyzed bacteriologically. and only 2 women had been pregnant Wve times (0. p = 0.8%) 90 (32.6%) 2 (1.01 0–5 1. Furthermore.564). p = 0.2% vs.06 0–4 1. and women with two deliveries 16.78 0–3 14 (10.5%) with two deliveries.5% of the women had three deliveries and in group 2 + 3 it was 3. p = 0.1%) 83 (30.9%) 1 (0.) Parity (p = n.0% vs.6%. 0. and in group 3.6%. with the same treatment for 2 weeks. Concomitant vaginal infections were found in 8.429).2%) 46 (16.3 18–50 29.5%) 1 (0. as well. whereat 4 suVered from vaginal trichomoniasis.3%). 9 patients were aVected by concomitant infections.5%). In group 1. 47. 136 women (32. and 10 from candidosis. 39. n = 162) were nulligravida.1% vs.1) 0. 7. No signiWcant statistical diVerence was observed in the distribution of Wndings between the three patient groups ( 2-test. p = 0. %) 0.6% vs. The analysis of the groups showed that 61. As much as 12 patients in the metronidazole group were aVected.878) nor between all three examined patient groups (ANOVA. 87 women (21. 152 women (36. The majority of women (39. as well as with four in 0. %) Age Mean (years) Range (years) Gravity Mean (n) Range (n) Parity Mean (n) Range (n) Mean (n.5) 0.0) 136 (32. The rate of deliveries was comparable in the diVerent groups (1 vs.Arch Gynecol Obstet (2010) 282:43–47 Table 1 Characteristics of all patients (n = 413) examined in the follow-up visit Total No. By the comparison of women treated with metronidazole (group 1) versus women treated with octenidine hydrochloride/phenoxyethanol (groups 2 + 3). no signiWcant statistical diVerence was observed in the distribution of the patients with respect to age groups ( 2-test.0%). 68 women (16.9) n.3%). 2 + 3. As shown in Table 2. see Table 2): the majority was nullipara (47.2%).0% of women were analyzed BV negative.8% patients in the group treated with octenidine hydrochloride/phenoxyethanol.7) Group 3 138 (33. p = 0. of patients (n.9%.3%) 41 (30.8% vs. Fig.5%) 55 (19.7%) 0 (0. 14 days).06 0–5 29.595).882).5) 1.106).0 18–50 Group 1 Group 2 139 (33. 8 women (1.0%) 277 413 (100. In the third group.4% in the two treatment groups.2%) were tested negative ( 2-test. As much as 124 women (30%) have had one pregnancy.6%) 8 (2.3%) 46 (33.6% of women were tested negative. and only 1 woman with four (0.) deliveries Group 1 Groups 2 + 3 Group 1 Groups 2 + 3 0 1 2 3 4 5 Total 53 (39.5 18–50 29.s. women with one delivery amounted to 33. p = 0.5%). With two pregnancies there were 23. p = 0. 30. From all women there were 196 nullipara (47.9%) four pregnancies. No signiWcant statistical diVerence was observed in the distribution of the number of pregnancies (Kruskal–Wallis test.9%) with three deliveries. 3 of them had vaginal trichomoniasis and 6 had candidosis. 1. p = 0. 71. 2 women suVered from vaginal trichomoniasis. see Table 1).6%) 0 (0.75 0–3 9 (6.5% patients in the group treated with metronidazole and 19.76 0–4 35 (8.

and presence of concomitant infections. and trichomoniasis is common. in 20–25% of cases vulvovaginal candidosis. Bacterial vaginal infectious diseases (bacterial colpitis. p = 0. especially in pregnancy. estrogen levels. Relapse rates reach around 60% within 3 months despite successful treatment [19]. parity.46 Arch Gynecol Obstet (2010) 282:43–47 Fig.027).8 and 4. and pathogenic bacteria the natural Xora is maintained stable. and a higher number of lifetime sexual partners [17].8% BV positive). groups 2 + 3: n = 277) after 14 days of therapy in comparison to the 7-day therapy regime ( 2-test. and that side eVects (increased secretion during the day) were lesser compared to the use of metronidazole vaginal tablets. Its onset has been associated with a change of sexual partner and vaginal douching [16]. metabolic products of the Xora. regarding the duration of therapy with octenidine hydrochloride/phenoxyethanol a statistically signiWcant positive eVect was observable: the percentage of women without BV at the end of therapy was signiWcantly higher in patients treated with a 14-days vaginal application regimen. 39. glycogen levels. Considering the fact that the groups treated with metronidazole vaginal tablets and locally applied octenidine hydrochloride/phenoxyethanol were homogenous with regards to the distribution of the patients’ age.2. It should be taken into consideration 123 . the physiological pH-value within the vagina is balanced between 3. Whether vaginosis is the result of an imbalance in the vaginal ecosystem or is sexually transmitted is unclear [15]. 13] and other endogenous bacteria. This might be explained by the persistence of Gardnerella vaginalis after clinical cure [20]. In general. and in 15–20% trichomoniasis [14]. Due to the ascending nature.7% in groups 2 + 3 (Fig. of which the most important for a healthy vagina is Lactobacillus crispatus [12. 3). the results have shown that bacterial vaginosis is treated eVectively by applying a local therapy with both metronidazole and octenidine hydrochloride/phenoxyethanol.. preterm delivery. for e.0% of the women were still aVected from BV in group 1 and 35. Bacterial vaginosis may remit and relapse. 36. most patients stated that octenidine hydrochloride/phenoxyethanol was more comfortable and easier to apply. Especially. In the follow-up visit. SigniWcantly more women were cured of bacterial vaginosis after the corresponding therapy (63. 2 Results of control smear analyses after the therapy of patients (n = 413) Fig. persist for several months (usually remitting after repeated treatments). Discussion Under healthy conditions the vaginal Xora is in a natural equilibrium and sustained through a complex mechanism of diVerent factors. these infections are associated with a variety of diVerent risks. Between the two treatment regimes no statistically signiWcant diVerence regarding the occurrence of BV was observed ( 2-test. gravidity. gonorrhoea. In the separate analysis of metronidazole treatment (group 1) versus therapy with octenidine hydrochloride/ phenoxyethanol (groups 2 + 3). Risk factors include use of an intra-uterine contraceptive device. which are found on the vaginal mucosa and known to be resistant to metronidazole treatment [21]. Co-infection with chlamydia. patients were asked by the attending physicians verbally about their satisfaction with the therapy as well as with the applied method.g. and bacterial vaginosis) are characterized by a disorder of the vaginal microbiological system with the elimination of the physiological lactobacilliXora. or resolve spontaneously [18]. p = 0. as well as of a persistence of dense and active bacterial bioWlm primarily consisting of Gardnerella vaginalis and Atopobium vaginae. early age of Wrst intercourse.2% BV negative vs. In general. As a result of an interaction between Lactobacillus species. in particular by a mixed Xora of pathogenic anaerobic bacteria. bacterial Fluor vaginalis. The present prospective study was aimed to evaluate the eVectiveness of diVerent therapies in the treatment of bacterial vaginosis.595). 3 Results of control smear analyses after therapy by treatment group (group 1: n = 136. In about 40–50% of all vaginal infections a bacterial vaginosis is diagnosed.

Eschenbach DA (1993) Bacterial vaginosis and anaerobes in obstetric-gynecologic infection.gov/std/treatment/ 2006/vaginal-discharge. Verstraelen H (2008) An adherent Gardnerella vaginalis bioWlm persists on the vaginal epithelium after standard therapy with oral metronidazole. Am J Obstet Gynecol 158:935–939 16. Mathis D. BMC Microbiol 7:115–128 13. Lochs H. Platz-Christensen J-J. Padian N. Marrazzo JM (2005) Molecular identiWcation of bacteria associated with bacterial vaginosis.o. Lipsky MS (2000) Diagnosis of vaginitis. J Infect Dis 167:783–784 20. Sobel JD. Hillier SL. Accessed 26 Apr 2009 11. the arrangement of a longer follow-up period was diYcult due to the character of an observational study. http:// www. Totten PA. Int J STD AIDS 8:603–608 19. ConXict of interest statement None. Burton JP. Friese K. Bump RC. Obstet Gynecol 77:450–452 4.gov/summary/summary. for donation of the necessary supply of octenidine hydrochloride/phenoxyethanol. and that the side eVects (increased secretion during the day) were lesser than with vaginal tablets. Swidsinski A. Holmes KK (1983) NonspeciWc vaginitis. Ferris DG. Pahlson C (1991) Clue cells in predicting infections after abdominal hysterectomy. Buesching WJ III (1988) Bacterial vaginosis in virginal and sexually active adolescent females: evidence against exclusive sexual transmission. Alqumber MA. and MIOFARM export–import d. and clindamycin vaginal cream. Forsum U (2005) Bacterial vaginosis—a disturbed bacterial Xora and treatment enigma. Litaker MS. Loening-Baucke V.htm#vagdis2. Hendrich J (1995) Treatment of bacterial vaginosis: a comparison of oral metronidazole. Chen KC. Am J Obstet Gynecol 198(1):97e1–97e6 22. Accessed 22 Jan 2009 2. Schmitt C. In: Vaginal discharge—sexually transmitted diseases. Verhelst R. Gardnerella vaginalis and Atopobium vaginae indicates an inverse relationship between L.aspx?doc_id=11602. Furthermore. Sex Trans Infect 80(1):8–11 6. Ugwumadu A.D.cdc. Chowns J (1997) Sex. but the patients stated that it was more comfortable and easier to apply. Obstet Gynecol 85:387–390 123 . Fiedler TL. Kirschner W (2000) Randomized trial of two local antiseptics in bacterial vaginal infections. Diagnostic criteria and microbial and epidemiologic associations. Am J Obstet Gynecol 169:446–469 3. Mendling W. Swidsinski S. Forsum U. women treated with the spray felt subjectively more contented. APMIS 113:305–316 that the follow-up period from 7 to 14 days after treatment might be too short to reXect the true cure rates. DoerVel Y. http://www. therapy with octenidine hydrochloride/ phenoxyethanol was as eVective as the therapy with the local antibiotic metronidazole. Clin Infect Dis 16(Suppl 4):S282–S287 5. Larsson PG. In conclusion. Sundström E (1991) Is bacterial vaginosis a sexually transmitted disease? Int J STD AIDS 2:362–364 18. because the application of the spray was more easy and comfortable in use. Meriwether C (1993) Long-term follow-up of patients with bacterial vaginosis treated with oral metronidazole and topical clindamycin. Schwabke JR (2003) Gynecologic consequences of bacterial vaginosis. Landers DV (1995) Plasma cell endometritis in women with symptomatic bacterial vaginosis. Eschenbach D. Hawes SE. Spiegel CA. However. N Engl J Med 353(18):1899–1911 14. References 1. Obstet Gynecol North Am 188(3):752–758 8. Mead PB (1993) Epidemiology of bacterial vaginosis. Vaneechoutte M (2007) Quantitative determination by real-time PCR of four vaginal Lactobacillus species. metronidazole vaginal gel. because asymptomatic patients considered themselves often as cured and were lost to follow-up. Germany) for her assistance in preparing the manuscript for publication. N Engl J Med 337(26):1896–1903 15. J Fam Pract 41:443–449 21. (Bonn. Sobel JD (1997) Vaginitis. Centers for Disease Control and Prevention (2006) Bacterial vaginosis. Siebert J. gasseri and L. Am J Med 74(1):14–22 10.guidelines. Temmerman M. iners. Bolan G. Tagg JR. Fredricks DN. De Backer E. Geburtsh Frauenheilk 60:308–313 12. Larsson P-G. thrush and bacterial vaginosis. Woodward L. and a longer period might be needed [22]. Acknowledgments We thank Andrea Rathmann-Schmitz. Wilson J (2004) Managing recurrent bacterial vaginosis. Am Fam Physician 62(5):1095–1104 9. Ohm-Smith M. Verstraelen H. Tenacious stains in clothes were diminished by using the colorless spray as well. Harke HP. Treatment Guidelines 2006. J Infect Dis 174:1058–1063 17.o.Arch Gynecol Obstet (2010) 282:43–47 47 7. Amsel R. Hay PE. Schachter J. Benedetti J et al (1996) Hydrogen peroxideproducing lactobacilli and acquisition of vaginal infections. Korn AP. Ph. Larsson PG. British Association for Sexual Health and HIV (2006) National Guideline for the Management of Bacterial Vaginosis. Platz-Christensen JJ. Neumann G. Egan ME.

or email articles for individual use. users may print.Copyright of Archives of Gynecology & Obstetrics is the property of Springer Science & Business Media B. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However. download. .V.