Article Text


How to treat persistent vaginal yeast infection due to species other than Candida albicans
  1. Sophia Davies1,
  2. Elizabeth Johnson2,
  3. David White1
  1. 1Department of Sexual Health and HIV, Heartlands Hospital, Birmingham, UK
  2. 2HPA South West Laboratory, HPA Mycology Reference Laboratory and National Collection of Pathogenic Fungi, Bristol, UK
  1. Correspondence to David White, Department of Sexual Health and HIV, Heartlands Hospital, Birmingham B95SS, UK; david.white{at}

Statistics from


Vulvovaginal candidiasis (VVC) is a common condition and usually straightforward to treat. In contrast, complicated VVC can be intractable and cause considerable psychological morbidity. Complicated VVC includes recurrent or severe disease, or when there are adverse factors in the host.1 This includes persistent infection with species other than Candida albicans (‘non-C albicans’) and the more common recurrent albicans VVC. The importance of distinguishing the two conditions is that non-C albicans chronic vaginal yeast infection is potentially completely curable, but may need a different approach in terms of treatment modalities.2–7 This article suggests a stepwise approach to treatment using the best current evidence and the clinical experience of the authors and focuses on C glabrata since this species is responsible for the majority of cases. Other species and indeed other genera such as Saccharomyces cerevisiae can also be involved but less frequently (see supplementary table 1).

Vaginal infection with non-C albicans may occur at any age and series from India and elsewhere suggest that uncontrolled type 2 diabetes is a risk factor.8 Symptoms may sometimes stretch back over many years with a long history of misdiagnosis and frustration. At least one swab identified to species level is an essential part of the diagnostic workup of any woman presenting with chronic or persistent vaginitis. Usually, but not always, the yeast is present in large amounts and is visible on microscopy. In the event of a positive swab a repeat sample should be sent to confirm. Relatively asymptomatic cases occur and clinically it can sometimes be difficult to be sure how much of a contributor the organism is to symptoms. This is sometimes only clear after eradication and even then symptoms can be very slow to settle.


By the time patients present with a non-C albicans infection they have almost by definition failed conventional therapy with azoles and nystatin (which can now only be obtained as an unlicensed import). Further treatment with conventional doses of oral or vaginal azoles may induce further resistance. However, if there is clinical urgency with a C glabrata infection, and there has not been extensive azole exposure then while results are awaited a reasonable choice is either nystatin or, if not available, miconazole nitrate (Gyno-Daktarin) 1200 mg on alternate days as isolates often have a lower minimum inhibitory concentration (MIC) to this agent than others in the class (see supplementary table 2) (E Johnson, unpublished data). This can be used with oral itraconazole 200 mg daily for 2 weeks. Otherwise treatment involves expensive agents and/or unlicensed preparations which may take considerable time to obtain. We regard these as ‘salvage’ treatments for which the aim is eradication of the organism from the vagina; for this reason prolonged treatment of at least 2 weeks has been necessary in reported series.3–7 There are however no randomised studies and the optimum length of treatment is unclear.

Salvage treatment

The characteristics that are important for successful salvage treatment are not known. One approach may be to use oral systemic fungicidal therapy in combination with topical therapy to penetrate into vaginal epithelia. It should be acknowledged that there are no established breakpoints for most of the topical antifungal agents. Moreover, accepted antifungal susceptibility breakpoints applied to drugs intended for systemic use cannot necessarily be applied to those agents when used topically as much higher local concentrations of the drugs may be achieved using topical treatment or at vaginal pH.9 Supplementary table 2 shows the susceptibility data for the 35 C glabrata isolates from vaginal swabs referred to the MRL during 2011 compared with the 204 C albicans isolates and as you can see the MIC50 results (ie, the MIC at which 50% of the isolates are inhibited) obtained for the azole drugs are often considerably higher for C glabrata than those for C albicans which usually remain easier to treat with azole therapy.

Our clinical experience tells us that although relapses are common with inadequate treatment, once finally eradicated the organism rarely returns (although attacks with other species may still occur). Species identification is therefore essential for any further attacks to reassure the patient that their condition has not recurred. Certain symptoms, particularly vulval soreness, may take time to resolve and cure should be based on the results of swabs taken at least a week after treatment. If swabs are negative but symptoms persist then patients should be reassured that they will probably, in the absence of any other vulvovaginal condition, settle with the use of local emollients (recommended for all patients) and the judicious use of local steroids. Our suggestion is to use aqueous cream applied to dry skin as a soap substitute and a greasy preparation, for example, Diprobase as a skin ‘waterproofer’.

Step 1 The best evidence for first-line salvage treatment centres on the use of intravaginal flucytosine (5FC) for 2–3 weeks.2–5 ,7 Based on the susceptibility profiles of C glabrata (see supplementary table 2) and the possibility of potential synergies between the two agents a product was formulated in the UK using 5FC in combination with amphotericin. Although 5FC is fungicidal, emergent resistance is quite common when it is used as monotherapy and primary resistance is seen with some species but it is uncommon with isolates of C glabrata. Whether this justifies initial susceptibility testing is a matter of debate. Polyene antifungals (particularly amphotericin) have been shown to be synergistic in their effect with regards to C glabrata and also with other fungi, for example, Cryptococcus.10 The reported series with amphotericin and 5FC in lubricating jelly have suggested 100%3 ,7 success but we have had personal experience of failure and have been contacted for advice about apparent failures. There have recently been problems with the supply of pure amphotericin powder, so the preparation has been reformulated with the substitution of nystatin (also a polyene) in place of amphotericin, and is available from the University of North Staffordshire Manufacturing Pharmacy.

Step 2 Following the failure of 5FC/nystatin the next logical agent to choose is intravaginal boric acid 600 mg for 2–3 weeks. This is available from Martindale Pharma, Brentwood, Essex. Series suggest a cure rate of 64–72%.4 ,6 Boric acid is safe when used for short courses, although long-term data are lacking.11

Step 3 If the condition fails to respond despite the use of these treatments the next step is unclear and clinical reporting is lacking. If the organism remains susceptible and there was a temporary response to treatment then it might be worth trying a prolonged course, that is, 4 weeks of either 5FC/nystatin or boric acid but there are currently no reports to support this.

Step 4 If the patient still has persistent infection the next step may depend on the susceptibility profile of the isolate. C glabrata will usually remain susceptible to 5FC and moderately susceptible to azoles. For C glabrata, miconazole is the most active in vitro of the vaginal preparations but it is unclear whether resistance profiles are truly helpful in defining therapy. Assuming the organism remains 5FC susceptible then it seems logical that this should be used intravaginally in any further treatment. If there is 5FC resistance it then makes sense to use either boric acid or vaginal pessaries, for example, clotrimazole 500 mg or miconazole nitrate (Gyno-Daktarin) 1200 mg on alternate days alongside intensive oral treatment. If azole therapy must be revisited then there remains a question as to whether this should be with voriconazole, posaconazole or high-dose fluconazole. Recent literature on the use of fluconazole 1200 mg in AIDS-related cryptococcal meningitis suggests that there is synergism with other agents and an improved rate of sterilisation with high doses, although that may reflect penetration into the cerebral spinal fluid so is not necessarily relevant to mucosal tissue infections.12 The unpredictable bioavailability, high cost and unique side-effect profile of voriconazole limit its use in this setting (the first two points also apply to posaconazole). Again the length of course is unclear but in practical terms this may be dictated by the patient's ability to tolerate the drug.

Step 5 If the patient remains infected then they may be effectively incurable. Suppressive vaginal boric acid may be a useful option here (although this will become very expensive, see table 1). Some patients have reported making their own preparations by putting boric acid into gelatine capsules but safety concerns preclude any recommendation to this course of action.

Table 1

Cost of various therapies used in the treatment of non-albicans VVC (2 weeks unless stated)

The future

Non-C albicans vaginal yeast infection is clearly a condition for which current evidence remains sparse, as is the case for most rare conditions. As such, this guideline largely comprises the expert opinion of the authors. Randomised trials are not feasible in view of the low numbers. We therefore encourage clinicians to download and complete the surveillance form at so that this information can be used to inform patients more accurately about the safety and success of treatment regimens.

Key messages

  • Chronic vaginal yeast infection with Candida species other than albicans is a rare but potentially completely curable condition.

  • The best evidence for first-line salvage treatment centres on the use of intravaginal flucytosine (5FC) in combination with amphotericin for 2–3 weeks.

  • Nystatin replaces amphotericin in the currently available formulation due to supply problems.

  • Other options include intravaginal boric acid and high-dose oral fluconazole or voriconazole.


View Abstract
  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • Contributors SD and DJW worked in collaboration to write and revise the manuscript. EJ provided data and expertise in several key areas.

  • Competing interests None.

  • Provenance and peer review Commissioned; externally peer reviewed.

  • Correction notice This article has been corrected since it was published Online First. The website has been corrected to

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.