brand logo

Am Fam Physician. 2000;61(11):3306-3312

See related patient information handout on recurrent yeast infections, written by the author of this article.

Vulvovaginal candidiasis is considered recurrent when at least four specific episodes occur in one year or at least three episodes unrelated to antibiotic therapy occur within one year. Although greater than 50 percent of women more than 25 years of age develop vulvovaginal candidiasis at some time, fewer than 5 percent of these women experience recurrences. Clinical evaluation of recurrent episodes is essential. Patients who self-diagnose may miss other causes or concurrent infections. Known etiologies of recurrent vulvovaginal candidiasis include treatment-resistant Candida species other than Candida albicans, frequent antibiotic therapy, contraceptive use, compromise of the immune system, sexual activity and hyperglycemia. If microscopic examination of vaginal secretions in a potassium hydroxide preparation is negative but clinical suspicion is high, fungal cultures should be obtained. After the acute episode has been treated, subsequent prophylaxis (maintenance therapy) is important. Because many patients experience recurrences once prophylaxis is discontinued, long-term therapy may be warranted. Patients are more likely to comply when antifungal therapy is administered orally, but oral treatment carries a greater potential for systemic toxicity and drug interactions.

More than 50 percent of women older than 25 years have one episode of vulvovaginal candidiasis,1 but fewer than 5 percent of these women experience recurrent infection.2 Vulvovaginal candidiasis is considered recurrent when at least four discrete episodes occur in one year or at least three episodes occur in one year and are not related to antibiotic therapy. Recurrent vulvovaginal candidiasis is distinguished from persistent infection by the presence of a symptom-free interval.


Women who have persistent or recurrent vulvovaginal candidiasis often present to their family physician with intense vaginal discomfort. Other presenting symptoms may include an odorless vaginal discharge, pruritus, dyspareunia or dysuria. Frequently, these women express their frustration with ineffective treatments.

Although the initial infection is sometimes diagnosed over the telephone, clinical evaluation of recurrent episodes is essential. Patients who self-diagnose yeast infections risk missing other etiologies or concurrent infections involving two or more organisms that require different treatments.

On physical examination, the patient with vulvovaginal candidiasis usually has vulvar erythema and a thick, white to yellow discharge in the vaginal vault.

Potassium hydroxide (KOH) testing and fungal cultures have some limitations, but they are still useful for identifying the infecting species and guiding treatment. Microscopic examination of vaginal secretions in a 10 percent KOH preparation may demonstrate hyphae. Characteristic budding mycelia are seen in fewer than 30 percent of positive candidal cultures.3 Fungal cultures should be obtained when clinical suspicion is high and the KOH preparation is negative, or when a patient has persistent or recurrent symptoms despite treatment. In addition, a wet mount should be examined for evidence of coexisting trichomoniasis or bacterial vaginosis.

Causes of Recurrence

Although Candida albicans is the pathogen identified in most patients with vulvovaginal candidiasis, other possible pathogens include Candida tropicalis and Candida glabrata. Increasingly, Candida species other than C. albicans have been found to cause yeast vaginitis (i.e., 9.9 percent of cases in 1988 and 17.2 percent of cases in 1995).4 In fact, recurrent infections may be caused by the resistance of non– C. albicans species to antifungal agents.

In vitro studies have shown that imidazole antifungal agents such as miconazole and clotrimazole are not as effective against non– C. albicans fungi. C. tropicalis and C. glabrata are 10 times less sensitive to miconazole than is C. albicans.5 Imidazoles are still the first-line treatment for C. albicans infections. A 1993 in vitro study examined more than 250 strains of C. albicans and found that no strain was resistant to ketoconazole, itraconazole and clotrimazole.6

Although antifungal resistance can cause treatment failure, other factors may contribute to recurrent vulvovaginal candidiasis (Table 1).5,715 Noncompliance with a treatment regimen may result in persistent infection that is mislabeled as a recurrence. For example, a patient may not complete the entire course of antifungal therapy, especially if an inconvenient topical treatment has been prescribed. A recurrence may also represent an inadequately treated infection. Between 15 and 20 percent of women with negative cultures after treatment have positive cultures within three months.16 If an infection recurs at least three months after the previous episode, it is more likely to be caused by a different C. albicans strain.7 Possible Risk Factors for Recurrent Vulvovaginal Candidiasis

Candida species
Diabetes mellitus
Mechanical irritation of vulvovaginal area
Inadequate treatment
Sexual transmission

Information from references 5 and 7 through 15.

Antibiotics are often implicated as a cause of recurrent vulvovaginal candidiasis. Frequent antibiotic use decreases protective vaginal flora and allows colonization by Candida species.8 The risk of a yeast infection increases with the duration of antibiotic use, but no specific antibiotic has been shown to be more likely to cause yeast infections.9

Diabetes mellitus is often considered a predisposing factor for recurrent vulvovaginal candidiasis. Hyperglycemia enhances the ability of C. albicans to bind to vaginal epithelial cells.10 However, unless other symptoms are suggestive of diabetes, patients with recurrent vulvovaginal candidiasis are rarely found to be diabetic.8

Contraceptive methods may also promote recurrences of vulvovaginal candidiasis. Use of spermicidal jellies and creams increases susceptibility to infection by altering the vaginal flora and increasing the adhesion of Candida organisms. Women who take oral contraceptive pills have a higher rate of vulvovaginal candidiasis.11 According to one theory, Candida cells have estrogen and progesterone receptors that, when stimulated, increase fungal proliferation.8

Women who are prone to recurrent vulvovaginal candidiasis may have deficient cell-mediated immunity. Similarly, persons with acquired immunodeficiency syndrome are susceptible to systemic candidal infection. Some studies suggest that 40 to 70 percent of women with recurrent vulvovaginal candidiasis have some specific anergy resulting in a subnormal T-lymphocyte response to Candida.12 One study found that Lewis A and B blood group antigens on the vaginal epithelium are protective against candidal infection.13

Mechanical factors may also be important. Perspiration associated with tightly fitted clothes or poorly ventilated underwear increases local temperature and moisture. Mechanical irritation of the vulvovaginal area by clothing or with sexual intercourse may also predispose already colonized areas to infection. One study demonstrated a positive relationship between the monthly frequency of sexual intercourse and the incidence of recurrent vulvovaginal candidiasis.14

Dietary habits have been suggested as causes of recurrent vulvovaginal candidiasis. However, most studies do not support a role for dietary factors in the etiology of recurrences, and adherence to strict diets has not been beneficial.8

The role of sexual transmission is controversial. One study found identical Candida strains in the sexual partners of 48 percent of women with recurrent infections.7 A randomized, controlled trial evaluated the effect that treating male sexual partners with oral ketoconazole had on the recurrence rates for vulvovaginal candidiasis.15 The recurrence rates in the treated and untreated partner groups were found to be similar at six months and one year. Topical antifungal therapy has been ineffective in male sexual partners, probably because of the presence of reservoirs not reached by this treatment. In summary, no clinical trial has found that the treatment of male sexual partners prevents recurrences of vulvovaginal candidiasis in women.

Some investigators have advocated the elimination of Candida from the gastrointestinal tract. The rationale is that reinfection from an intestinal reservoir contributes to vaginal recurrences. However, studies have not found an association between recurrent vulvovaginal candidiasis and the presence of intestinal Candida.16


The optimal treatment for recurrent vulvovaginal candidiasis has not yet been defined (Table 2). Consequently, treatment must be individualized based on a comparison of effectiveness, convenience, potential side effects and cost.

If a patient has infrequent recurrences, the simplest and most cost-effective regimen involves self-diagnosis and the early initiation of topical therapy. A prospective randomized, open, crossover study in 23 women with proven recurrent vulvovaginal candidiasis examined the efficacy and cost benefit of monthly prophylaxis compared with empiric self-treatment at the onset of symptoms.17 Patients were randomized to receive one 500-mg dose of clotrimazole intravaginally each month with the menses for six months or one 500-mg dose of clotrimazole intravaginally at the onset of symptoms. After six months, patients were switched (crossover) to the other regimen.

In this study,17 50 episodes of symptomatic vaginitis (2.2 episodes per patient) occurred with the prophylactic regimen, and 86 episodes (3.7 episodes per patient) occurred with empiric treatment. During the prophylactic period, the women used an average of 7.3 doses of clotrimazole, compared with 3.6 doses during the empiric treatment period. When asked about their personal preference, almost 75 percent of the women preferred the empiric regimen. The authors of the study concluded that although prophylactic perimenstrual clotrimazole therapy may reduce the number of symptomatic episodes, empiric self-treatment is more cost-effective and acceptable to patients. Problems with the empiric regimen include inappropriate use and a delay in diagnosis if the patient does not have vaginal candidiasis. In addition, the 500-mg troche of clotrimazole is no longer available.

If a woman with an established diagnosis of recurrent vulvovaginal candidiasis does not respond to an imidazole, infection with a resistant non– C. albicans species may be present. Terconazole vaginal cream (Terazol) is the agent of choice when infection with a species other than C. albicans is suspected. The potent interference of this agent with the cytochrome P450 isoenzymes makes C. tropicalis and C. glabrata more susceptible to treatment.

After the acute episode of recurrent vulvovaginal candidiasis has been treated, subsequent prophylaxis or maintenance therapy is essential. In one clinical trial, women with a history of recurrent vulvovaginal candidiasis were randomized to receive 400 mg of ketoconazole for 14 days or clotrimazole in the form of 100-mg vaginal suppositories for seven days.18 One week after treatment, the clinical cure rate was higher than 80 percent in both groups. Two months after treatment, in the absence of any maintenance therapy, 53 percent of women in the ketoconazole treatment group and 63 percent of those in the clotrimazole treatment group had recurrences.

Several maintenance regimens have been studied. In one clinical trial, 74 women with recurrent vulvovaginal candidiasis were treated for an acute episode with 400 mg of ketoconazole per day for 14 days.19 The women were then randomized to receive one of three treatments: placebo, 400 mg of ketoconazole administered orally for five days after the menses for six months, or 100 mg of ketoconazole administered orally each day for six months. The six-month recurrence rates were 71 percent for the placebo group, 29 percent for the cyclic-regimen group and 5 percent for the daily-regimen group.

Maintenance therapy needs to be given frequently enough to prevent vaginal regrowth, but the optimal dosing interval is not clear. One study suggested that the weekly administration of 0.8 percent terconazole vaginal cream is nearly as effective as daily treatment with ketoconazole.20 Similar efficacy has been noted for twice-weekly intravaginal treatment with 200 mg of clotrimazole.21

A monthly 150-mg dose of orally administered fluconazole has been shown to reduce the incidence of recurrences by 50 percent.22 Itraconazole, in a dosage of 200 mg23 or 400 mg24 administered orally once a month, also has been found to decrease the recurrence rate by approximately 50 percent. Boric acid, administered in a 600-mg vaginal suppository twice daily for two weeks and then daily during menstruation, has been effective in the treatment of women with resistant infection.3 However, the use of boric acid is limited by significant local irritation and the possibility of intoxication.25

Based on the study findings, ketoconazole (Nizoral) administered orally once a day, clotrimazole (Gyne-Lotrimin) administered intravaginally twice weekly, terconazole administered intravaginally once a week, and fluconazole (Diflucan) or itraconazole (Sporanox) administered orally once a month have been relatively effective in reducing the recurrence rate for vulvovaginal candidiasis.

Most studies recommend prophylaxis for six months. Then the woman is reevaluated. Many women have recurrences once prophylaxis is discontinued. Thus, they may need to stay on medication for a longer period.

The expense of each regimen should be considered. The costs given in Table 2 do not include the expenses associated with decreased work productivity, missed work days, toxicity monitoring or office visits. Treatment Options for Recurrent Vulvovaginal Candidiasis

AgentDosing regimenCost for brand name (generic) regimen*
Treatment of acute episode
Clotrimazole (Gyne-Lotrimin)100-mg tablets administered intravaginally for seven days$ 16
Terconazole 0.8 percent cream (Terazol 3)One full applicator (5 g) administered intravaginally for three days29
Fluconazole (Diflucan)150 mg administered orally (one dose)12
Ketoconazole (Nizoral)200 mg administered orally once daily for 14 days47 (42)
400 mg administered orally once daily for 14 days94 (84)
Boric acid600-mg vaginal suppository administered twice daily for 14 daysNA
Prophylaxis (maintenance)
Clotrimazole (Gyne-Lotrimin)Two 100-mg tablets administered intravaginally twice weekly for six months32 §
Ketoconazole (Nizoral)Two 200-mg tablets administered orally for five days after the menses for six months33 (30)§
One half of a 200-mg tablet administered orally once daily for six months51 (46)§
Terconazole 0.8 percent cream (Terazol 3)One full applicator (5 g) administered vaginally once a week29§
Fluconazole (Diflucan)150 mg administered orally once a month12§
Itraconazole (Sporanox)One 200-mg tablet administered orally once a month7§
Two 200-mg tablets administered orally once a month14§
Boric acid600-mg vaginal suppository administered once daily during menstruation (5-day menses)NA

NA = not available commercially.

*—Estimated cost to the pharmacist based on average wholesale prices (rounded to the nearest dollar) in Red book. Montvale, N.J.: Medical Economics Data, 1999. Cost to the patient will be greater, depending on prescription filling fee.

—Dosages based on the literature, not on indications given in Physicians' Desk Reference. 53d ed. Montvale, N.J.: Medical Economics, 1999.

—Although boric acid suppositories are not commerically available, they can be compounded; based on figures from the University of Missouri, the cost for compounding one suppository ranges from $0.75 to $1.00.

§—Monthly cost.

A 1992 crossover study assessed the association between the daily ingestion of yogurt containing Lactobacillus acidophilus and the prevention of recurrent vulvovaginal candidiasis.26 Women were assigned to a yogurt-free diet or a yogurt-containing diet. Although only 13 of 21 women completed the protocol, the women who ingested yogurt had a threefold reduction in infection. The authors of the study concluded that daily ingestion of 8 oz of yogurt containing L. acidophilus decreased the rate of candidal infection. A second study showed no difference in infection rates between women who ingested pasteurized yogurt and women who ingested yogurt that contained L. acidophilus.27 Additional evidence is needed before management recommendations can be made.

In addition to cost, other factors may determine the most appropriate regimen. Compliance rates are greater for medications that are taken orally rather than intravaginally. However, the potential for systemic toxicity and drug interactions is greater with orally administered medications.

Gastrointestinal side effects occur in 15 percent of patients treated with orally administered antifungal agents.28 In addition, hepatic toxicity has been noted in one of 15,000 persons treated with orally administered ketoconazole.29 Although clotrimazole therapy may cause local discomfort, it is less frequently associated with systemic toxicity (headache occurs in 9 percent of recipients and abdominal pain is a problem in 3 percent of recipients).30

Compared with ketoconazole, fluconazole is less likely to be toxic. Because fluconazole is administered orally, treatment compliance is better than with clotrimazole, which is administered intravaginally. Patients treated with fluconazole report headache (12 percent), abdominal pain (7 percent) and nausea (4 percent).30

Many antifungal agents have significant interactions with other medications. For example, interactions have been noted between fluconazole and warfarin (Coumadin), oral hypoglycemic agents, phenytoin (Dilantin), theophylline and rifampin (Rifadin).31 Other drugs reported to interact with fluconazole include cyclosporine (Sandimmune), zidovudine (Retrovir) and hydrochlorothiazide (Esidrix).

Continue Reading

More in AFP

More in PubMed

Copyright © 2000 by the American Academy of Family Physicians.

This content is owned by the AAFP. A person viewing it online may make one printout of the material and may use that printout only for his or her personal, non-commercial reference. This material may not otherwise be downloaded, copied, printed, stored, transmitted or reproduced in any medium, whether now known or later invented, except as authorized in writing by the AAFP.  See permissions for copyright questions and/or permission requests.