ACOG

PRACTICE
BULLETIN
CLINICAL MANAGEMENT GUIDELINES FOR
OBSTETRICIAN–GYNECOLOGISTS
NUMBER 72, MAY 2006

Vaginitis
Vaginal symptoms are common in the general population and are one of the
This Practice Bulletin was most frequent reasons for patient visits to obstetrician–gynecologists (1).
developed by the ACOG Com- Vaginitis may have important consequences in terms of discomfort and pain,
mittee on Practice Bulletins—
days lost from school or work, and sexual functioning and self-image. Vaginitis
Gynecology with the assistance
is associated with sexually transmitted diseases and other infections of the
of Paul Nyirjesy, MD. The in-
formation is designed to aid female genital tract, including human immunodeficiency virus (HIV) (2, 3), as
practitioners in making deci- well as adverse reproductive outcomes in pregnant and nonpregnant women.
sions about appropriate obstet- Treatment usually is directed to the specific causes of vaginal symptoms, which
ric and gynecologic care. These most commonly include bacterial vaginosis, vulvovaginal candidiasis, and tri-
guidelines should not be con- chomoniasis. The purpose of this document is to provide information about the
strued as dictating an exclusive diagnosis and treatment of vaginitis.
course of treatment or proce-
dure. Variations in practice may
be warranted based on the
needs of the individual patient,
Background
resources, and limitations Vaginitis is defined as the spectrum of conditions that cause vulvovaginal
unique to the institution or type symptoms such as itching, burning, irritation, and abnormal discharge. The
of practice. most common causes of vaginitis are bacterial vaginosis (22–50% of sympto-
matic women), vulvovaginal candidiasis (17–39%), and trichomoniasis
Reaffirmed 2019 (4–35%); 7–72% of women with vaginitis may remain undiagnosed (4). In the
undiagnosed group of women, symptoms may be caused by a broad array of
conditions, including atrophic vaginitis, various vulvar dermatologic condi-
tions, and vulvodynia. Vaginitis has a broad differential diagnosis, and success-
ful treatment frequently rests on an accurate diagnosis.
Estrogen status plays a crucial role in determining the normal state of the
vagina. In the prepubertal and postmenopausal states, the vaginal epithelium is
thinned, and the pH of the vagina usually is elevated (4.7 or greater). A routine
bacterial culture will demonstrate a broad variety of organisms, including skin
and fecal flora. During the reproductive years, the presence of estrogen increas-
es glycogen content in vaginal epithelial cells, which in turn encourages colo-
nization of the vagina by lactobacilli. This increased level of colonization leads
to lactic acid production and a resulting decrease in the either 1) visualization of blastospores or pseudohyphae
vaginal pH to less than 4.7. However, even in women of on saline or 10% KOH microscopy, or 2) a positive cul-
reproductive age, the normal vaginal flora remain hetero- ture in a symptomatic woman. The diagnosis can be
geneous, and other components of the vaginal flora, such further classified as uncomplicated or complicated vul-
as Gardnerella vaginalis, Escherichia coli, group B vovaginal candidiasis (see the box). This classification
streptococci (GBS), genital mycoplasmatales, and Can- system has treatment implications because complicated
dida albicans, are commonly found. vulvovaginal candidiasis is more likely to fail standard
Evaluation of women with vaginitis should include a antifungal therapy (5, 6).
focused history about the entire spectrum of vaginal Women with uncomplicated vulvovaginal candidia-
symptoms, including change in discharge, vaginal mal- sis can be treated successfully with any of the options in
odor, itching, irritation, burning, swelling, dyspareunia, Table 1. Topical treatments may cause local side effects,
and dysuria. Questions about the location of symptoms such as burning and irritation. Occasionally, oral therapy
(vulva, vagina, anus), duration, the relation to the men- may cause systemic side effects, such as gastrointestinal
strual cycle, the response to prior treatment including intolerance, headache, and liver function test elevations;
self-treatment and douching, and a sexual history can these usually are mild and self-limited (5). Allergic reac-
yield important insights into the likely etiology. Because tions to oral therapy are rare. Because all listed antifun-
many patients with vaginitis have vulvar manifestations gal treatments seem to have comparable safety and
of disease, the physical examination should begin with a efficacy, the choice of therapy should be individualized to
thorough evaluation of the vulva. However, evaluation the specific patient; factors such as cost, convenience,
may be compromised by patient self-treatment with non- compliance, ease of use, history of response or adverse
prescription medications. During speculum examination, reactions to prior treatments, and patient preference can
samples should be obtained for vaginal pH, amine all be taken into consideration.
(“whiff”) test, and saline (wet mount) and 10% potassi- Patients with complicated vulvovaginal candidiasis
um hydroxide (KOH) microscopy. The pH and amine require more aggressive treatment to achieve relief of
testing can be performed either through direct measure- symptoms. In a placebo-controlled randomized trial of
ment or by colorimetric testing. It is important that the women with severe vulvovaginal candidiasis, a second
swab for pH evaluations be obtained from the mid-por- dose of fluconazole, 150 mg given 3 days after the first
tion of the vaginal side wall to avoid false elevations in dose, increased the cure rate from 67% to 80% (6). In
pH results caused by cervical mucus, blood, semen, or
other substances. In selected patients, vaginal cultures or
polymerase chain reaction tests for trichomonas or yeast
are helpful. A vaginal Gram stain for Nugent scoring of Classification of Vulvovaginal Candidiasis
the bacterial flora may help to identify patients with bac- Uncomplicated
terial vaginosis. Other currently available ancillary tests Sporadic or infrequent episodes
for diagnosing vaginal infections include rapid tests for
Mild to moderate symptoms or findings
enzyme activity from bacterial vaginosis-associated
organisms, Trichomonas vaginalis antigen, and point-of- Suspected Candida albicans infection
care testing for DNA of G vaginalis, T vaginalis, and Nonpregnant woman without medical complica-
Candida species; however, the role of these tests in the tions
proper management of patients with vaginitis is unclear. Complicated
Depending on risk factors, DNA amplification tests can Recurrent episodes (four or more per year)
be obtained for Neisseria gonorrheae and Chlamydia tra-
Severe symptoms or findings
chomatis.
Suspected or proved non-albicans Candida
Vulvovaginal Candidiasis infection
Women with diabetes, severe medical illness,
Physical manifestations of vulvovaginal candidiasis immunosuppression, other vulvovaginal conditions
range from asymptomatic colonization to severely symp-
tomatic. Symptomatic women may report itching, burn- Pregnancy
ing, irritation, dyspareunia, burning with urination, and a Modified from Sexually transmitted diseases treatment guidelines
whitish thick discharge. Multiple studies conclude that a 2002. Centers for Disease Control and Prevention. MMWR Recomm
Rep 2002;51(RR-6):1–78.
reliable diagnosis cannot be made on the basis of history
and physical examination alone (4). Diagnosis requires

2 ACOG Practice Bulletin No. 72

Table 1. Therapy for Vulvovaginal Infections (Drugs Listed Alphabetically)

Indication Drug Formulation Dosage Duration

Uncomplicated Butoconazole 2% sustained-release cream 5 g daily 1 day
vulvovaginal candidiasis Clotrimazole 1% cream 5 g daily 7 days
2% cream 5 g daily 3 days
100-mg vaginal suppository 100 mg daily 7 days
200-mg vaginal suppository 200 mg daily 3 days
500-mg vaginal suppository 500 mg daily 1 day
Fluconazole 150-mg oral tablet 150 mg daily 1 day
Miconazole 2% cream 5 g daily 7 days
100-mg vaginal suppository 100 mg daily 7 days
200-mg vaginal suppository 200 mg daily 3 days
1,200-mg vaginal suppository 1,200 mg daily 1 day
Nystatin 100,000 units vaginal tablets daily 14 days
Terconazole 0.4% cream 5 g daily 7 days
0.8% cream 5 g daily 3 days
Tioconazole 2% cream 5 g daily 3 days
6.5% cream 5 g daily 1 day
Bacterial vaginosis Clindamycin 2% cream 5 g daily 7 days
2% sustained-release cream 5 g daily 1 day
100-mg ovules 100 mg daily 3 days
300-mg oral 300 mg twice daily 7 days
Metronidazole 0.75% gel 5 g daily 5 days
500-mg oral 500 mg twice daily 7 days
Trichomoniasis Metronidazole 500-mg oral 4 tabs as one dose 1 day
500 mg twice daily 7 days
Tinidazole 500-mg oral 4 tabs as one dose 1 day
Data from Sexually transmitted diseases treatment guidelines 2002. Centers for Disease Control and Prevention. MMWR Recomm Rep 2002;51(RR-6):1–78; Sobel JD,
Faro S, Force RW, Foxman B, Ledger WJ, Nyirjesy P, et al. Vulvovaginal candidiasis: epidemiologic, diagnostic, and therapeutic considerations. Am J Obstet Gynecol
1998;178:203–11; Cohen L. Treatment of vaginal candidosis using clotrimazole vaginal cream: single dose versus 3-day therapy. Curr Med Res Opin 1985;9:520–3;
Faro S, Skokos CK. The efficacy and safety of a single dose of Clindesse vaginal cream versus a seven-dose regimen of Cleocin vaginal cream in patients with bacterial
vaginosis. Clindesse Investigators Group. Infect Dis Obstet Gynecol 2005;13:155–60; Gabriel G, Robertson E, Thin RN. Single dose treatment of trichomoniasis. J Int
Med Res 1982;10:129–30.

women with recurrent vulvovaginal candidiasis second- mg weekly or 200 mg twice a week, are acceptable
ary to C albicans, after initial intensive therapy for 7–14 options (9). Candida species colonization and sympto-
days to achieve mycologic remission, prolonged antifun- matic vulvovaginal candidiasis may occur more com-
gal treatment with fluconazole, 150 mg weekly (7) for 6 monly in pregnant women (10). Although low-dose
months, will successfully control more than 90% of short-term fluconazole use is not associated with known
symptomatic episodes and will lead to a prolonged pro- birth defects (11), higher doses of 400–800 mg/d have
tective effect in approximately 50% of women. Although been linked to birth defects (12). Thus, treatment of vul-
daily oral ketoconazole was previously described as an vovaginal candidiasis in pregnancy should consist of one
effective suppressive therapy in women with recurrent of the topical imidazole therapies listed in Table 1, prob-
vulvovaginal candidiasis (8), weekly fluconazole has a ably for 7 days (13).
lower risk of liver toxicity and should be used instead of Although much less common than C albicans, vul-
ketoconazole (9). For patients who are unable or unwill- vovaginal candidiasis caused by non-albicans Candida
ing to take fluconazole, prolonged maintenance therapy species are less likely to respond to azole antifungal
with intermittent topical agents, such as clotrimazole, 500 therapy (6). Current experience consists exclusively of

ACOG Practice Bulletin No. 72 3

descriptions of case series of patients seen at centers spe- Following treatment, bacterial vaginosis may recur
cializing in the treatment of vaginitis. A standard course in up to 30% of women within 3 months (24). Possible
of topical imidazole therapy may be effective in up to mechanisms include persistence of pathogenic bacteria,
50% of such cases (14). Therapy with vaginal boric acid, reinfection from exogenous sources, including a sexual
600-mg capsules daily for a minimum of 14 days, seems partner, or failure of the normal lactobacillus-dominant
to be effective for azole failures (15). Patients with non- flora to reestablish themselves. Studies of partner treat-
albicans Candida vulvovaginal candidiasis in whom ment have failed to show a protective effect (25, 26).
boric acid therapy is ineffective should be referred to a Studies of recolonization with lactobacillus supplements
specialist experienced in handling such cases. have used nonvaginal strains of lactobacillus and have
failed to show a clear benefit (24). Prolonged antibiotic
Bacterial Vaginosis therapy may be useful in women with recurrent bacterial
Bacterial vaginosis is a polymicrobial infection marked vaginosis; however, further investigation is warranted.
by a lack of hydrogen peroxide-producing lactobacilli Nonpregnant women with bacterial vaginosis can be
and an overgrowth of facultative anaerobic organisms. treated with the alternatives listed in Table 1. Although
Organisms that are found with greater frequency and clindamycin use may be associated with in vitro anti-
numbers in women with bacterial vaginosis include microbial resistance (21), the listed alternatives seem to
G vaginalis, Mycoplasma hominis, Bacteroides species, have comparable clinical efficacy and safety (27–29).
Peptostreptococcus species, Fusobacterium species, Generally, topical therapy is more expensive than gener-
Prevotella species, Atopobium vaginae, and other anaer- ic oral metronidazole, although the latter may be asso-
obes (16, 17). Because these organisms are part of the ciated with significant gastrointestinal symptoms.
normal flora, the mere presence of these organisms, espe- Disulfiramlike reactions may occur with both oral and
cially G vaginalis, on a culture does not mean that the topical metronidazole. As with the treatment of vulvo-
patient has bacterial vaginosis. Patients with bacterial vaginal candidiasis, treatment for BV should be individ-
vaginosis, when symptomatic, may complain of an ualized to the patient after considering multiple clinical
abnormal vaginal discharge and a fishy odor. A clinical factors.
diagnosis of bacterial vaginosis requires the presence of In several epidemiologic studies, bacterial vaginosis
three out of four Amsel’s criteria: abnormal gray dis- has been associated with low birth weight, premature
charge, vaginal pH greater than 4.5, a positive amine test, rupture of membranes (PROM), and prematurity
and more than 20% of the epithelial cells being clue cells. (30–32). Standard antibiotic therapy seems to effectively
In research settings, the Nugent score (18), which assigns eradicate bacterial vaginosis in pregnant women (33, 34),
a value to different bacterial morphotypes seen on Gram and those with symptomatic bacterial vaginosis should be
stain of vaginal secretions, is considered the current cri- treated. Neither metronidazole nor clindamycin have
terion standard for diagnosing bacterial vaginosis. known teratogenic effects (35). Studies have been con-
Compared to Nugent scoring, Amsel’s criteria have a sen- ducted to determine whether treating asymptomatic bac-
sitivity of 92% and specificity of 77% (19). However, a terial vaginosis in an uncomplicated pregnancy will
similar sensitivity and specificity have been demonstrat- decrease the risk of adverse outcomes. They have yield-
ed by using any combination of only two clinical criteria ed conflicting results and have shown no clear benefit to
(20). routine screening and treatment in U.S. populations (36).
In nonpregnant women, bacterial vaginosis has been However, in women with high-risk pregnancies, particu-
associated with a number of infections of the female larly those with prior preterm deliveries, some studies
reproductive tract, including pelvic inflammatory disease have shown that screening for and treating bacterial vagi-
(PID), postprocedural gynecologic infections, and acqui- nosis with oral metronidazole may decrease the risk of
sition of HIV and herpes simplex virus (HSV)-2 infec- preterm PROM and preterm delivery (37, 38), but others
tions (21). Treatment for bacterial vaginosis before have not (36).
abortion or hysterectomy significantly decreases the risk
of postoperative infectious complications (22). Treatment Trichomoniasis
helps women to resolve concurrent mucopurulent cer- Vaginal trichomoniasis is a common sexually transmitted
vicitis (23). There are no current data on the treatment of disease with an estimated annual incidence of 7.4 million
bacterial vaginosis to decrease acquisition of PID, HIV, cases in the United States (39). Symptomatic women
or HSV-2, and the role of treatment of asymptomatic bac- with trichomoniasis may have an abnormal discharge,
terial vaginosis to prevent these associated morbidities is itching, burning, or postcoital bleeding. Although many
unclear. women with trichomoniasis will have an elevated vaginal

4 ACOG Practice Bulletin No. 72

pH, diagnosis in clinical settings relies on visualization Other Causes of Vaginal Symptoms
of motile trichomonads on saline microscopy. A wet
Although vulvovaginal candidiasis, bacterial vaginosis,
mount has a sensitivity of 55–60% in diagnosing tri-
and trichomoniasis cause the most vulvovaginal symp-
chomoniasis (40, 41). Trichomonas culture techniques
toms, other causes may include a broad range of condi-
are associated with greater than 90% sensitivity (42). A
tions such as vulvar diseases, atrophic vaginitis, and rarer
point-of-care test for trichomonas antigens, the OSOM
forms of vaginitis.
Trichomonas Rapid Test, has a sensitivity of 88.3% and
Patients with atrophic vaginitis may have an abnor-
specificity of 98.8% (43) compared with culture. This
mal vaginal discharge, dryness, itching, burning, or dys-
test may be a valuable diagnostic tool, particularly in set-
pareunia. Although more common in postmenopausal
tings with a high prevalence of trichomoniasis and where
women, sometimes it can be observed in younger pre-
microscopy or culture is not available.
menopausal women. Diagnosis can be made on the basis
Treatments for uncomplicated trichomoniasis are
of an elevated vaginal pH and the presence of parabasal
listed in Table 1. Although metronidazole has been the
or intermediate cells on microscopy. An amine test result
mainstay of treatment in the United States, tinidazole
will be negative. Treatment consists of local water-based
also has recently been approved as single-dose therapy.
Both treatments seem to be equally efficacious (9). Side moisturizing preparations or topical or systemic estrogen
effects seem to be of similar nature, including a possible (48).
disulfiramlike effect; alcohol should be avoided for 24 Of the rarer forms of vaginitis, the best defined
hours after metronidazole use and 72 hours after tinida- seems to be desquamative inflammatory vaginitis (49).
zole use (9). Partners of women with trichomoniasis also Generally occurring in perimenopausal or postmeno-
should be treated. In cases of metronidazole allergy, pausal women, desquamative inflammatory vaginitis
patients may be referred for desensitization to and treat- causes burning, dyspareunia, and an abnormal yellow or
ment with metronidazole. There are no data on cross- green discharge. Although streptococcal species, includ-
reactivity between tinidazole and metronidazole. ing GBS, are found in more than 90% of affected women,
Metronidazole is considered safe to use in preg- this does not mean that desquamative inflammatory
nancy; data on tinidazole are too limited to be of use. vaginitis is caused by streptococcal species. Some have
Like bacterial vaginosis, trichomoniasis has been associ- argued that desquamative inflammatory vaginitis may
ated with adverse outcomes such as preterm delivery, represent a vaginal expression of erosive lichen planus
PROM, and low birth weight (42). Although a study of (50). Examination reveals a purulent discharge with vary-
treatment for asymptomatic trichomoniasis in pregnant ing amounts of vestibular and vaginal erythema. The
women showed an increased preterm delivery rate in the vaginal pH is elevated and the amine test result is nega-
treated group, it should be noted that 23% of the women tive. Microscopy reveals large amounts of polymor-
in the placebo group received metronidazole outside the phonuclear cells and parabasal cells. This condition is
protocol, and treatment occurred at advanced gestational easily mistaken for trichomoniasis; however, in cases of
age. Therefore, these results may not apply to a broader desquamative inflammatory vaginitis, no motile tri-
population of pregnant women (44). However, the results chomonads are present and cultures for T vaginalis are
of this study suggest that treatment of trichomoniasis negative. Although no randomized controlled studies
during pregnancy does not help to prevent associated have been performed, a 14-day course with 2% clin-
adverse sequelae. damycin cream often will achieve a cure; however,
Although high-level resistance to metronidazole is relapse after therapy is fairly common (49).
considered rare, low level in vitro resistance may be as
high as 5% (45). In suspected cases of resistance, patients
should be interviewed carefully to exclude the possibility Clinical Considerations and
of noncompliance with the medication regimen or rein-
fection from an untreated partner. In a series of 33 cases, Recommendations
treatment with high-dosage tinidazole, 500 mg four times
▲

When are vaginal cultures helpful in making

daily or more for 14 days, was well tolerated and effective the diagnosis of vaginitis?
in more than 90% of resistant cases (46). A lower dosage
of tinidazole, 500 mg three times daily for 7 days, was Although microscopy is considered the standard in clini-
also effective in a series of three resistant cases (47). cal practice, its sensitivity to yeast is around 50% and it
Sending the resistant isolate to a reference laboratory that misses a substantial percentage of patients with sympto-
can perform susceptibility testing should be considered to matic vulvovaginal candidiasis (51, 52). Self-treatment
help guide choice and dosing of therapy (9). before evaluation also may make it more difficult for the

ACOG Practice Bulletin No. 72 5

health care provider to visualize yeast on microscopy. result, is frequently isolated in women with vaginal
Furthermore, compared with culture and yeast poly- symptoms as well. However, a case–control study of 118
merase chain reaction, false-positive rates of up to 50% women with GBS found no association between women
have been reported (53). Because they can pick up small- with GBS and vulvovaginal symptoms (56). Similarly,
er numbers of organisms, yeast cultures are considered lactobacilli are part of the vaginal flora. Although it has
the criterion standard in confirming the presence of yeast. been hypothesized that an overgrowth of lactobacilli can
They are not routinely performed because of their cost, cause vaginal symptoms (57), such a syndrome is poor-
the delay involved in obtaining results, and the fact that ly characterized, and controlled studies confirming the
many women may be asymptomatically colonized with existence of such a syndrome are lacking. Thus, the pres-
yeast. Nevertheless, yeast cultures should be obtained in ence of large numbers of lactobacilli on either
cases of recurrent vulvovaginal candidiasis or possible microscopy or vaginal culture should be considered a
non-albicans Candida infection; the latter should be sus- normal finding.
pected if microscopy reveals only blastospores or the

▲
patient with vulvovaginal candidiasis has persistent When is it appropriate to provide treatment
symptoms after antifungal therapy. Yeast cultures also for vaginitis without an examination?
should be considered in symptomatic women with nega-
tive microscopy, those with signs of vulvovaginal can- Over the past decade, women have increasingly relied on
didiasis, or multiple symptoms but negative microscopy self-diagnosis and self-treatment of vulvovaginal can-
results (54). didiasis. An estimated $275 million is spent annually on
Because microscopy has a fairly limited sensitivity, nonprescription antifungals, and the drugs number in the
culture or trichomonas antigen testing should be obtained top 10 of all nonprescription medications sold in the
in situations where trichomoniasis is suspected but not United States (58). With topical antimycotic agents
proved. However, health care providers may have diffi- approved for nonprescription use, it is assumed that
culty finding a laboratory that can provide a culture women with a prior episode of vulvovaginal candidiasis
medium and perform the test. There are currently no clear can self-diagnose accurately (59). The perceived benefits
criteria or studies to assess which patients should under- of nonprescription antifungals include convenience, the
go trichomonas cultures. Their use should be considered ability to rapidly initiate antimycotic therapy, and the
in patients with a negative wet mount test result and any potential to reduce health care costs significantly (1).
of the following circumstances: a history of trichomonia- However, the reliability of self-diagnosis may be
sis with persistent symptoms after therapy, a high vaginal poorer than previously suggested. In a study of 601
pH and microscopy that reveals leukocytes, a Pap test women recruited from a variety of medical and commu-
result with trichomonas, or patient desire for trichomonas nity sites in Georgia, investigators found that only 11%
screening because of a possible exposure. of women with no prior diagnosis and 34.5% of women
Mucopurulent cervicitis, which is sometimes caused with a prior diagnosis of vulvovaginal candidiasis accu-
by Neisseria gonorrheae or C trachomatis (55), may rately recognize the classic scenario for candidiasis (60).
present as an abnormal yellow discharge. Therefore, Both groups were particularly poor at recognizing bacte-
DNA tests or cultures for these two organisms should be rial vaginosis, with an accuracy of 3.2% and 4.4%,
obtained in patients with a purulent discharge, cervical respectively. In a prospective study of 95 symptomatic
friability, any symptoms suggestive of PID, or leukocytes women purchasing nonprescription antifungal products,
on microscopy. Such tests also should be performed in only 34% had pure vulvovaginal candidiasis, and self-
women who fall into higher risk groups where annual treatment with a topical antifungal agent would have
screening is recommended (9). been inappropriate or inadequate therapy in the remain-
Because the normal vaginal flora is very heteroge- ing 66% (61). In a longitudinal study of women who
neous, routine bacterial cultures of the vagina have no submitted yeast cultures every 4 months for a year,
use in diagnosing bacterial vaginosis. They may have a researchers found no correlation between antecedent
limited role in diagnosing suspected cases of group A Candida species colonization and subsequent antifungal
streptococcal vaginitis, but this condition is considered use (62). Finally, a telephone diagnosis of vaginal symp-
rare. In patients with symptoms suggestive of bacterial toms seemed to correlate poorly with the actual diagno-
vaginosis that do not fulfill Amsel’s criteria, a Gram stain sis (63). Given the nonspecific nature of vulvovaginal
is considered the criterion standard for diagnosis. Other symptoms (19), patients who are already in the office
organisms routinely found on vaginal culture include should not be treated for vaginitis without an examina-
GBS and lactobacilli. Group B streptococci is part of the tion. Whenever possible, patients requesting treatment
normal flora in approximately 25% of women and, as a by telephone should be asked to come in for evaluation;

6 ACOG Practice Bulletin No. 72

this is particularly true of a woman who has treated her- treatment may compromise the ability of her health care
self with a nonprescription antifungal and still has per- provider to obtain an accurate diagnosis and will further
sistent symptoms. However, in a known compliant delay the ability to initiate appropriate therapy.
patient with multiple confirmed prior episodes who Furthermore, women who use numerous courses of non-
reports the same symptoms as before, a short course of prescription antifungal therapy and do not have vulvo-
treatment can be initiated over the phone. If she fails to vaginal candidiasis may incur significant financial costs.
improve, she should be asked to come in for evaluation. Because many of the adverse effects of nonprescription
products are secondary to an imprecise diagnosis, accu-
▲

How should patients be evaluated in the rate home tests for vaginitis ultimately may help to min-
absence of a microscope? imize these effects.
There may be times when patients can only be evaluated

▲
For symptomatic patients with a high pH but
without microscopy. Because there are currently no rapid
tests for yeast, testing for vulvovaginal candidiasis with- normal microscopy, what is appropriate man-
out a microscope will consist of history, examination, agement?
and culture. An elevated vaginal pH will determine Testing of the vaginal pH and amine testing are part of a
which patients may need further testing for bacterial battery of tests that are used to diagnose vulvovaginal
vaginosis or trichomoniasis. Testing for trichomoniasis symptoms. When pH is abnormally elevated in a symp-
can be performed with point-of-care tests for tri- tomatic patient, it is usually associated with microscopic
chomonas antigen (the OSOM Trichomonas Rapid Test) findings that help to establish a diagnosis. Depending on
or culture. Point-of-care tests for pH and amines the cause of symptoms, findings such as trichomonads,
(QuickVue Advance pH and Amines test), G vaginalis clue cells, or immature epithelial cells may be seen.
proline iminopeptidase activity (QuickVue Advance G.
However, recent intercourse, menses, sampling of cervi-
vaginalis test) and vaginal sialidases (OSOM BVBlue
cal mucus, or recent treatment with a medication also
test) are all FDA-approved to aid in the diagnosis of bac-
can alter the pH of the vagina. In the presence of com-
terial vaginosis. Although their exact role in current
pletely normal microscopy (including vaginal cytology),
diagnostic algorithms is unclear, their use should be con-
there is no evidence that a high pH alone causes vaginal
sidered when a microscope is unavailable. When possi-
symptoms. Thus, the symptomatic patient should be
ble, a slide of vaginal secretions should be obtained for
treated in a manner similar to other women with vagini-
future Gram stain.
tis where the diagnosis is unclear, including obtaining
cultures for yeast and trichomonas.
▲

Are there adverse effects of nonprescription

antifungal use?
▲

For findings of bacterial vaginosis or

In general, topical nonprescription antifungal use is trichomonas on a cytology report, what is
associated with cure rates and side effects that are simi- appropriate management?
lar to prescription therapy. A patient with vulvovaginal
candidiasis who uses a nonprescription product should The Pap test is an unreliable tool for diagnosing either
respond to therapy; failure to respond to initial treatment bacterial vaginosis or trichomoniasis. When compared to
should prompt clinical evaluation. Physical side effects Gram stain criteria for bacterial vaginosis, a Pap test has
consist primarily of localized burning and irritation in a sensitivity of 49% and specificity of 93% (65). In
about 5% of women (5). If used for the wrong condition symptomatic women with bacterial vaginosis on a Pap
or if the patient has vulvovaginal candidiasis but fails test, a vaginal pH, amine test, and wet mount should be
to respond to treatment, antifungal use may lead to a performed; asymptomatic women do not need evaluation
delay in accurate diagnosis and appropriate treatment. or treatment given that the diagnosis on Pap test is uncer-
Although such delay may have minimal effect on vulvo- tain and it is unclear that asymptomatic nonpregnant
vaginal symptoms, such as itching, odor, or discharge, it women with bacterial vaginosis benefit from treatment
may be of greater concern if a patient who self-treats for (9). For trichomoniasis, the Pap test has a sensitivity sim-
vulvovaginal candidiasis actually has PID, a sexually ilar to the wet mount but yields a false-positive rate of at
transmitted infection, or a urinary tract infection (61). least 8% with standard tests (40) and 4% with liquid-
Isolation of resistant Candida species as a result of wide- based cytology (66); thus, a diagnosis based on cytology
spread nonprescription antifungal use seems to be can lead to an inaccurate diagnosis of a sexually transmit-
uncommon (64). If a woman who self-treats with a non- ted infection. When feasible, in patients with trichomonas
prescription agent fails to improve with therapy, her found on a Pap test, a wet mount and, if negative, a cul-

ACOG Practice Bulletin No. 72 7

ture should be performed. If culture is unavailable, the In adolescent patients, the causes of vaginitis are
least expensive approach is to treat the patient with similar to an adult population of reproductive age (72).
metronidazole. In populations with a low prevalence of In sexually active adolescents with vaginitis, screening
trichomoniasis (5% or less), this approach may lead to for gonorrhea and chlamydia also should be performed.
unnecessary treatment in more than 50% of patients (67). In adolescents who wish to avoid a speculum examina-
tion, examination of swabs obtained blindly from the
▲

What nonmedical approaches are effective? vagina have a sensitivity similar to speculum examina-
tions (72) for diagnosing causes of vaginitis, and urine
Complementary and alternative therapies are commonly testing can be performed for gonorrhea and chlamydia if
used to treat vulvovaginal symptoms (68). Such thera- indicated.
pies include lactobacilli, yogurt, garlic, tea tree oil, a low
carbohydrate diet, desensitization to Candida species

▲
How should patients be counseled?
antigen, hormonal manipulation with depot medroxy-
progesterone, and douching. Current data are insufficient Several specific myths may need to be addressed in
regarding either efficacy or safety to support recommen- counseling patients about vaginitis. Following is a dis-
dation of these nonmedical treatments for bacterial vagi- cussion of some common questions that may arise dur-
nosis or vulvovaginal candidiasis (69). ing counseling:
• Which types of vaginitis are sexually transmitted
▲

For vaginitis in pediatric or adolescent diseases (STDs) and which are not? Did I get this
patients, what is appropriate management, from my current sexual partner? Trichomoniasis is
and are there any special considerations? an STD. However, because asymptomatic carriage
can occur for prolonged periods in both men and
Vulvovaginitis is one of the most common gynecologic women, a recent diagnosis of trichomoniasis does
problems in prepubertal girls. However, because of the not necessarily establish recent acquisition, unless
lack of estrogenization of the vagina and resulting vagi- the patient has had documented negative tri-
nal atrophy and alkalinic pH, the causes seem to be quite chomonas cultures in the past. Because men can
different from an adult population. Most cases are thought harbor T vaginalis, a woman with trichomoniasis
to be noninfectious in origin, secondary to a broad range should refrain from intercourse until both she and
of conditions, many of them dermatologic (eg, contact her partner(s) have been treated. Although bacterial
dermatitis). Those cases with specific bacterial causes vaginosis is associated with sexual activity (73), it
typically have an acute onset of a visible discharge. also has been described in virginal women (74) and
Respiratory organisms such as group A streptococci is not considered an STD. However, in female part-
and Hemophilus influenzae are the most common infec- ners of lesbians with bacterial vaginosis, there is a
tious causes (70), as well as enteric and sexually trans- higher incidence of bacterial vaginosis (75); no
mitted pathogens; Candida species is rarely found. Lichen studies address whether simultaneous treatment of
sclerosis and atrophic vaginitis also may be present in both women in a lesbian couple will decrease recur-
prepubertal girls. Pinworms may cause perianal and vul- rence rates. Although vulvovaginal candidiasis also
var itching. A pediatric patient with vulvovaginal symp- is associated with sexual factors, such as oral recep-
toms should undergo a careful vulvar examination to look tive sex, it does not seem to be an STD (76). With
for evidence of a dermatologic cause and for vaginal dis- both bacterial vaginosis and vulvovaginal candidia-
charge. Vaginal secretions should be evaluated by sis in heterosexual couples, randomized studies of
microscopy to look for leukocytes (70), and a bacterial partner treatment have failed to show a decrease in
culture should be obtained by introducing a swab through the risk of recurrence (22, 77).
the hymen. Therapy depends on the results of the • What is the role of douching in the prevention or
microscopy and culture. An examination for pinworms treatment of vaginitis? No studies show any benefit
may demonstrate the presence of pinworm eggs. In cases to douching as a treatment for vaginitis. The associ-
of a possible foreign body, the discharge often will have ation of douching with bacterial vaginosis (73) and
an abnormal odor and be associated with some vaginal bacterial vaginosis–associated flora (16), although
bleeding. Vaginal irrigation may lead to expulsion of the not a clear demonstration of cause and effect, sug-
foreign body; if not, vaginoscopy should be performed. If gests that douching should not be used as a treat-
sexual abuse is suspected, child protective services ment for vaginitis and actually may exacerbate
should be notified and the child referred to a professional symptoms. In addition, douching has been associat-
trained in the management of such cases (71). ed with increased risk of cervicitis, PID, and tubal

8 ACOG Practice Bulletin No. 72

 infertility in retrospective studies (78). However, a References
recent prospective study failed to find an association
between douching, cervicitis, and PID (79). 1. Lipsky MS, Waters T, Sharp LK. Impact of vaginal anti-
fungal products on utilization of health care services: evi-
• What is the role of tampons in causing vaginitis? dence from physician visits. J Am Board Fam Pract
Tampon use does not seem to be associated with 2000;13:178–82. (Level II-3)
either bacterial vaginosis (73) or vulvovaginal can- 2. Sewankambo N, Gray RH, Wawer MJ, Paxton L,
didiasis (76). Women who are using an intravaginal McNaim D, Wabwire-Mangen F, et al. HIV-1 infection
product to treat a vaginal infection may want to associated with abnormal vaginal flora morphology and
avoid use of tampons during treatment to ensure bacterial vaginosis [published erratum appears in Lancet
adequate dispersion of the medication. 1997;350:1036]. Lancet 1997;350:546–50. (Level II-2)
3. Taha ET, Hoover DR, Dallabetta GA, Kumwedna NI,
Mtimavalye LA, Yang LP, et al. Bacterial vaginosis and
Summary of disturbances of vaginal flora: association with increased
acquisition of HIV. AIDS 1998;12:1699–706. (Level III)
Recommendations and 4. Anderson MR, Klink K, Cohrssen A. Evaluation of vagi-
nal complaints. JAMA 2004;291:1368–79. (Level III)
Conclusions
5. Sobel JD, Brooker D, Stein GE, Thomason JL, Wermeling
The following recommendations are based on DP, Bradley B, et al. Single oral dose fluconazole com-
good and consistent scientific evidence (Level A): pared with conventional clotrimazole topical therapy of
Candida vaginitis. Fluconazole Vaginitis Study Group.
Am J Obstet Gynecol 1995;172:1263–8. (Level I)
▲

Women with complicated vulvovaginal candidiasis

should receive more aggressive treatment than 6. Sobel JD, Kapernick PS, Zervos M, Reed BD, Hooton T,
women with an uncomplicated episode. Soper D, et al. Treatment of complicated Candida vagini-
tis: comparison of single and sequential doses of flucona-
▲

To prevent reinfection, women with trichomoniasis zole. Am J Obstet Gynecol 2001;185:363–9. (Level I)
should avoid intercourse until they and their partner
7. Sobel JD, Wiesenfeld HC, Martens M, Danna P, Hooton
have received treatment. TM, Rompalo A, et al. Maintenance fluconazole therapy
for recurrent vulvovaginal candidiasis. N Engl J Med
The following recommendations and conclusions 2004;351:876–83. (Level I)
are based on limited or inconsistent scientific evi- 8. Sobel JD. Recurrent vulvovaginal candidiasis. A prospec-
dence (Level B): tive study of the efficacy of maintenance ketoconazole
therapy. N Engl J Med 1986;315:1455–8. (Level I)
▲

Microscopy is the first line for diagnosing vulvo- 9. Sexually transmitted diseases treatment guidelines 2002.
vaginal candidiasis and trichomoniasis. In selected Centers for Disease Control and Prevention. MMWR
patients, culture for yeast and T vaginalis should be Recomm Rep 2002;51(RR-6):1–78.
obtained in addition to standard office-based testing. 10. Cotch MF, Hillier SL, Gibbs RS, Eschenbach DA.
Epidemiology and outcomes associated with moderate to
▲

Douching is not recommended for the prevention or

treatment of vaginitis. heavy Candida colonization during pregnancy. Vaginal
Infections and Prematurity Study Group. Am J Obstet
▲

Self-diagnosis of vaginitis is unreliable. Gynecol 1998;178:374–80. (Level II-2)

11. Sorensen HT, Nielsen GL, Olesen C, Larsen H, Steffensen
The following recommendation is based primarily FH, Schonheyder HC, et al. Risk of malformations and
on consensus and expert opinion (Level C): other outcomes in children exposed to fluconazole in
utero. Br J Clin Pharmacol 1999;48:234–8. (Level II-3)
▲

Clinical evaluation of women with vaginal symp-

12. Pursley TJ, Blomquist IK, Abraham J, Andersen HF,
toms should be encouraged, particularly for women Bartley JA. Fluconazole-induced congenital anomalies in
who fail to respond to self-treatment with a nonpre- three infants. Clin Infect Dis 1996;22:336–40. (Level III)
scription antifungal. 13. Xu J, Sobel JD. Candida vulvovaginitis in pregnancy.
Curr Infect Dis Rep 2004;6:445–9. (Level III)
14. Sood G, Nyirjesy P, Weitz MV, Chatwani A. Terconazole
Proposed Performance cream for non Candida albicans fungal vaginitis: results of
Measure a retrospective analysis. Infect Dis Obstet Gynecol
2000;8:240–3. (Level III)
Percentage of women with vulvovaginal candidiasis in 15. Sobel JD, Chaim W, Nagappan V, Leaman D. Treatment
whom diagnosis was confirmed with microscopy or culture of vaginitis caused by Candida glabrata: use of topical

ACOG Practice Bulletin No. 72 9

 boric acid and flucytosine. Am J Obstet Gynecol tion and with bacterial vaginosis. Obstet Gynecol
2003:189:1297–300. (Level III) 1986;67:229–37. (Level II-2)
16. Ness RB, Hillier SL, Richter HE, Soper DE, Stamm C, 31. Hillier SL, Nugent RP, Eschenbach DA, Krohn MA,
McGregor J, et al. Douching in relation to bacterial vagi- Gibbs RS, Martin DH, et al. Association between bacteri-
nosis, lactobacilli, and facultative bacteria in the vagina. al vaginosis and preterm delivery of low-birth-weight
Obstet Gynecol 2002;100:765–72. (Level II-2) infant. The Vaginal Infections and Prematurity Study
Group. N Engl J Med 1995;333:1737–42. (Level II-2)
17. Burton JP, Devillard E, Cadieux PA, Hammond JA, Reid
G. Detection of Atopobium vaginae in postmenopausal 32. Goldenberg RL, Hauth JC, Andrews WW. Intrauterine
women by cultivation-independent methods warrants fur- infection and preterm delivery. N Engl J Med 2000;
ther investigation. J Clin Microbiol 2003;42:1829–31. 342:1500–7. (Level III)
(Level III) 33. Yudin MH, Landers DV, Meyn L, Hillier SL. Clinical and
18. Nugent RP, Krohn MA, Hillier SL. Reliability of diag- cervical cytokine response to treatment with oral or vagi-
nosing bacterial vaginosis is improved by a standardized nal metronidazole for bacterial vaginosis during preg-
method of Gram stain interpretation. J Clin Microbiol nancy: a randomized trial. Obstet Gynecol 2003;102:
1991;29:297–301. (Level II-3) 527–34. (Level I)
19. Landers DV, Wiesenfeld HC, Heine RP, Krohn MA, 34. Ugwumadu A, Manyonda I, Reid F, Hay P. Effect of early
Hillier SL. Predictive value of the clinical diagnosis of oral clindamycin on late miscarriage and preterm delivery
lower genital tract infection in women. Am J Obstet in asymptomatic women with abnormal vaginal flora and
Gynecol 2004;190:1004–10. (Level II-3) bacterial vaginosis: a randomised controlled trial. Lancet
2003;361:983–7. (Level I)
20. Gutman RE, Peipert JF, Weitzen S, Blume J. Evaluation of
clinical methods for diagnosing bacterial vaginosis. 35. Diav-Citrin O, Shechtman S, Gotteiner T, Arnon J, Ornoy
Obstet Gynecol 2005;105:551–6. (Level II-2) A. Pregnancy outcome after gestational exposure to
metronidazole: a prospective controlled cohort study.
21. Beigi RH, Austin MN, Meyn LA, Krohn MA, Hillier SL.
Teratology 2001;63:186–92. (Level II-2)
Antimicrobial resistance associated with the treatment
of bacterial vaginosis. Am J Obstet Gynecol 2004;191: 36. Carey JC, Klebanoff MA, Hauth JC, Hillier SL, Thom
1124–9. (Level I) EA, Ernest JM, et al. Metronidazole to prevent preterm
delivery in pregnant women with asymptomatic bacterial
22. Koumans EH, Kendrick JS. Preventing adverse sequelae
vaginosis. National Institute of Child Health and Human
of bacterial vaginosis: a public health program and
Devleopment Network of Maternal-Fetal Medicine Units.
research agenda. CDC Bacterial Vaginosis Working
N Engl J Med 2000;342:534–40. (Level I)
Group. Sex Transm Dis 2001;28:292–7. (Level III)
37. Morales WJ, Schorr S, Albritton J. Effect of metronida-
23. Schwebke JR, Weiss HL. Interrelationships of bacterial
zole in patients with preterm birth in preceding pregnancy
vaginosis and cervical inflammation. Sex Transm Dis
and bacterial vaginosis: a placebo-controlled, double-
2002;29:59–64. (Level I)
blind study. Am J Obstet Gynecol 1994;171:345–7; dis-
24. Wilson J. Managing recurrent bacterial vaginosis. Sex cussion 348–9. (Level I)
Transm Infect 2004;80:8–11. (Level III)
38. Hauth JC, Goldenberg RL, Andrews WW, DuBard MB,
25. Vejtorp M, Bollerup AC, Vejtorp L, Fanoe E, Nathan E, Copper RL. Reduced incidence of preterm delivery with
Reiter A, et al. Bacterial vaginosis: a double-blind ran- metronidazole and erythromycin in women with bacterial
domized trial of the effect of treatment of the sexual part- vaginosis. N Engl J Med 1995;333:1732–6. (Level I)
ner. Br J Obstet Gynaecol 1988;95:920–6. (Level I)
39. Weinstock H, Berman S, Cates W Jr. Sexually transmitted
26. Colli E, Landoni M, Parazzini F. Treatment of male part- diseases among American youth: incidence and preva-
ners and recurrence of bacterial vaginosis: a randomised lence estimates, 2000. Perspect Sex Reprod Health 2004;
trial. Genitourin Med 1997;73:267–70. (Level I) 36:6–10. (Level III)
27. Ferris DG, Litaker MS, Woodward L, Mathis D, Hendrich 40. Krieger JN, Tam MR, Stevens CE, Nielsen IO, Hale J,
J. Treatment of bacterial vaginosis: a comparison of oral Kiviat NB, et al. Diagnosis of trichomoniasis. Compari-
metronidazole, metronidazole vaginal gel, and clin- son of conventional wet-mount examination with cytolog-
damycin vaginal cream. J Fam Pract 1995;41:443–9. ic studies, cultures, and monoclonal antibody staining of
(Level I) direct specimens. JAMA 1988;259:1223–7. (Level II-3)
28. Hanson JM, McGregor JA, Hillier SL, Eschenbach DA, 41. Pastorek JG 2nd, Cotch MF, Martin DH, Eschenbach DA.
Kreutner AK, Galask RP, et al. Metronidazole for bacter- Clinical and microbiological correlates of vaginal tri-
ial vaginosis. A comparison of vaginal gel vs. oral thera- chomoniasis during pregnancy. The Vaginal Infections
py. J Reprod Med 2000;45:889–96. (Level I) and Prematurity Study Group. Clin Infect Dis 1996;23:
29. Paavonen J, Mangioni C, Martin MA, Wajszczuk CP. 1075–80. (Level II-2)
Vaginal clindamycin and oral metronidazole for bacterial 42. Soper D. Trichomoniasis: under control or undercon-
vaginosis: a randomized trial. Obstet Gynecol 2000;96: trolled? Am J Obstet Gynecol 2004;190:281–90. (Level
256–60. (Level I) III)
30. Gravett MG, Hummel D, Eschenbach DA, Holmes KK. 43. Huppert JS, Batteiger BE, Braslins P, Feldman JA, Hobbs
Preterm labor associated with subclinical amniotic infec- MM, Sankey HZ, et al. Use of an immunochromato-

10 ACOG Practice Bulletin No. 72

 graphic assay for rapid detection of trichomonas vaginalis 60. Ferris DG, Dekle C, Litaker MS. Women’s use of over-
in vaginal specimens. J Clin Microbiol 2005;43:684–7. the-counter antifungal medications for gynecologic symp-
(Level II-3) toms. J Fam Pract 1996;42:595–600. (Level II-3)
44. Klebanoff MA, Carey JC, Hauth JC, Hillier SL, Nugent 61. Ferris DG, Nyirjesy P, Sobel JD, Soper D, Pavletic A,
RP, Thom EA, et al. Failure of metronidazole to prevent Litaker MS. Over-the-counter antifungal drug misuse
preterm delivery among pregnant women with asympto- associated with patient-diagnosed vulvovaginal candidia-
matic Trichomonas vaginalis infection. National Institute sis. Obstet Gynecol 2002;99:419–25. (Level II-3)
of Child Health and Human Development Network of 62. Beigi RH, Meyn LA, Moore DM, Krohn MA, Hillier SL.
Maternal-Fetal Medicine Units. N Engl J Med 2001;345: Vaginal yeast colonization in nonpregnant women: a lon-
487–93. (Level I) gitudinal study. Obstet Gynecol 2004;104:926–30. (Level
45. Schmid G, Narcisi E, Mosure D, Secor WE, Higgins J, II-2)
Moreno H. Prevalence of metronidazole-resistant Tricho- 63. Allen-Davis JT, Beck A, Parker R, Ellis J, Polley D.
monas vaginalis in a gynecology clinic. J Reprod Med Assessment of vulvovaginal complaints: accuracy of tele-
2001;46:545–9. (Level II-3) phone triage and in-office diagnosis. Obstet Gynecol
46. Sobel JD, Nyirjesy P, Brown W. Tinidazole therapy for 2002;99:18–22. (Level II-3)
metronidazole-resistant vaginal trichomoniasis. Clin 64. Mathema B, Cross E, Dun E, Park S, Bedell J, Slade B, et
Infect Dis 2001;33:1341–6. (Level III) al. Prevalence of vaginal colonization by drug-resistant
47. Hager WD. Treatment of metronidazole-resistant Tricho- Candida species in college-age women with previous
monas vaginalis with tinidazole: case reports of three exposure to over-the-counter azole antifungals. Clin Infect
patients. Sex Transm Dis 2004;31:34–-5. (Level III) Dis 2001;33:E23–7. (Level II-2)
48. Ballagh SA. Vaginal hormone therapy for urogenital and 65. Greene JF 3rd, Kuehl TJ, Allen SR. The papanicolaou
menopausal symptoms. Semin Reprod Med 2005;23: smear: inadequate screening test for bacterial vaginosis
126–40. (Level III) during pregnancy. Am J Obstet Gynecol 2000;182:
1048–9. (Level II-3)
49. Sobel JD. Desquamative inflammatory vaginitis: a new
subgroup of purulent vaginitis responsive to topical 2% 66. Lara-Torre E, Pinkerton JS. Accuracy of detection of tri-
clindamycin therapy. Am J Obstet Gynecol 1994;171: chomonas vaginalis organisms on a liquid-based papani-
1215–20. (Level III) colaou smear. Am J Obstet Gynecol 2003;188:354–6.
(Level II-3)
50. Edwards L, Friedrich EG Jr. Desquamative vaginitis:
lichen planus in disguise. Obstet Gynecol 1988;71:832–6. 67. Wiese W, Patel SR, Patel SC, Ohl CA, Estrada CA. A
(Level III) meta-analysis of the Papanicolaou smear and wet mount
for the diagnosis of vaginal trichomoniasis. Am J Med
51. Abbott J. Clinical and microscopic diagnosis of vaginal
2000;108:301–8. (Meta-analysis)
yeast infection: a prospective analysis. Ann Emerg Med
1995;25:587–91. (Level II-3) 68. Nyirjesy P, Weitz MV, Grody MH, Lorber B. Over-the-
counter and alternative medicines in the treatment of
52. Nyirjesy P, Seeney SM, Grody MH, Jordan CA, Buckley
chronic vaginal symptoms. Obstet Gynecol 1997;90:
HR. Chronic fungal vaginitis: the value of cultures. Am J
50–3. (Level II-3)
Obstet Gynecol 1995;173:820–3. (Level II-3)
69. Van Kessel K, Assefi N, Marrazzo J, Eckert L. Common
53. Ledger WJ, Polaneczky MM, Yih MC, Jeremias J, Tolbert
complementary and alternative therapies for yeast vagini-
V, Witkin SS. Difficulties in the diagnosis of candida
tis and bacterial vaginosis: a systematic review. Obstet
vaginitis. Infect Dis Clin Pract 2000;9:66–9. (Level III)
Gynecol Surv 2003;58:351–8. (Level III)
54. Eckert LO, Hawes SE, Stevens CE, Koutsky LA,
70. Stricker T, Navratil F, Sennhauser FH. Vulvovaginitis in
Eschenbach DA, Holmes KK. Vulvovaginal candidiasis:
prepubertal girls. Arch Dis Child 2003;88:324–6. (Level
clinical manifestations, risk factors, management algo-
III)
rithm. Obstet Gynecol 1998;92:757–65. (Level II-2)
71. Kass-Wolff JH, Wilson EE. Pediatric gynecology: assess-
55. Willmott FE. Mucopurulent cervicitis: a clinical entity? ment strategies and common problems. Semin Reprod
Genitourin Med 1988;64:169–71. (Level II-3) Med 2003;21:329–38. (Level III)
56. Shaw C, Mason M, Scoular A. Group B streptococcus car- 72. Blake DR, Duggan A, Quinn T, Zenilman J, Joffe A.
riage and vulvovaginal symptoms: causal or casual? A Evaluation of vaginal infections in adolescent women: can
case-control study in a GUM clinic population. Sex it be done without a speculum? Pediatrics 1998;
Transm Infect 2003;79:246–8. (Level II-2) 102:939–44. (Level II-3)
57. Cibley LJ, Cibley LJ. Cytolytic vaginosis. Am J Obstet 73. Hawes SE, Hillier SL, Benedetti J, Stevens CE, Koutsky
Gynecol 1991;165:1245–9. (Level III) LA, Wolner-Hanssen P, et al. Hydrogen peroxide-produc-
58. Marrazzo J. Vulvovaginal candidiasis. BMJ 2003;326: ing lactobacilli and acquisition of vaginal infections. J
993–4. (Level III) Infect Dis 1996;174:1058–63. (Level II-2)
59. Chaponis RJ, Bresnick PA, Weiss RR, Edwards LD. 74. Bump RC, Buesching WJ 3rd. Bacterial vaginosis in vir-
Candida vaginitis: signs and symptoms aid women’s self- ginal and sexually active adolescent females: evidence
recognition. J Clin Res Drug Dev 1993;7:17–23. (Level against exclusive sexual transmission. Am J Obstet
III) Gynecol 1988;158:935–9. (Level II-2)

ACOG Practice Bulletin No. 72 11

75. Marrazzo JM, Koutsky LA, Eschenbach DA, Agnew K,
Stine K, Hillier SL. Characterization of vaginal flora and The MEDLINE database, the Cochrane Library, and the
bacterial vaginosis in women who have sex with women. American College of Obstetricians and Gynecologists’ own
J Infect Dis 2002;185:1307–13. (Level II-2) internal resources and documents were used to conduct a
literature search to locate relevant articles published be-
76. Geiger AM, Foxman B. Risk factors in vulvovaginal can- tween January 1985 and February 2006. The search was re-
didiasis: a case-control study among university students. stricted to articles published in the English language.
Epidemiology 1996;7:182–7. (Level II-2) Priority was given to articles reporting results of original
77. Fong IW. The value of treating the sexual partners of research, although review articles and commentaries also
women with recurrent vaginal candidiasis with ketocona- were consulted. Abstracts of research presented at sympo-
zole. Genitourin Med 1992;68:174–6. (Level I) sia and scientific conferences were not considered adequate
for inclusion in this document. Guidelines published by or-
78. Scholes D, Stergachis A, Ichikawa LE, Heidrich FE,
ganizations or institutions such as the National Institutes of
Holmes KK, Stamm WE. Vaginal douching as a risk fac- Health and ACOG were reviewed, and additional studies
tor for cervical Chlamydia trachomatis infection. Obstet were located by reviewing bibliographies of identified arti-
Gynecol 1998;91:993–7. (Level II-3) cles. When reliable research was not available, expert opin-
79. Ness RB, Hillier SL, Kip KE, Richter HE, Soper DE, ions from obstetrician–gynecologists were used.
Stamm CA, et al. Douching, pelvic inflammatory disease, Studies were reviewed and evaluated for quality according
and incident gonococcal and chlamydial genital infection to the method outlined by the U.S. Preventive Services Task
in a cohort of high-risk women. Am J Epidemiol Force:
2005;161:186–95. (Level II-2)
I Evidence obtained from at least one properly de-
signed randomized controlled trial.
II-1 Evidence obtained from well-designed controlled
trials without randomization.
II-2 Evidence obtained from well-designed cohort or
case–control analytic studies, preferably from more
than one center or research group.
II-3 Evidence obtained from multiple time series with or
without the intervention. Dramatic results in uncon-
trolled experiments also could be regarded as this
type of evidence.
III Opinions of respected authorities, based on clinical
experience, descriptive studies, or reports of expert
committees.
Based on the highest level of evidence found in the data,
recommendations are provided and graded according to the
following categories:
Level A—Recommendations are based on good and consis-
tent scientific evidence.
Level B—Recommendations are based on limited or incon-
sistent scientific evidence.
Level C—Recommendations are based primarily on con-
sensus and expert opinion.

Copyright © May 2006 by the American College of Obstetricians and

Gynecologists. All rights reserved. No part of this publication may be
reproduced, stored in a retrieval system, posted on the Internet, or
transmitted, in any form or by any means, electronic, mechanical, pho-
tocopying, recording, or otherwise, without prior written permission
from the publisher.
Requests for authorization to make photocopies should be directed to
Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA
01923, (978) 750-8400.
ISSN 1099-3630
The American College of Obstetricians and Gynecologists
409 12th Street, SW, PO Box 96920, Washington, DC 20090-6920

Vaginitis. ACOG Practice Bulletin No. 72. American College of

Obstetricians and Gynecologists. Obstet Gynecol 2006;107:1195–1206.