Effectiveness of treating non-specific pruritus vulvae with topical steroids: A randomized controlled trial

Abstract

Objective: To determine the effectiveness of triamcinolone cream in the treatment of non-specific pruritus vulvae. Methods: A total of 50 patients, aged between 8 and 55 years, with non-specific pruritus vulvae were consecutively included in a double-blind randomized controlled trial by their general practitioner in 25 general practices in the Netherlands. Analysis was by “intention to treat”. An infectious cause was excluded by a vaginal and cervical swab. After randomization, 27 patients were treated with triamcinolone cream and 23 patients with a placebo cream. The effectiveness was assessed by: change in the severity of itch visual analogue scale (VAS) score; the percentage of patients with complete recovery; and changes in general health according to COOP/WONCA charts. Results: The mean decrease in severity of itch was 2.08 points in the triamcinolone group compared to 3.26 points in the placebo group. Forty-two per cent of the women in the triamcinolone group completely recovered versus 35% in the placebo group. There was no difference in general health perception between the two groups. After 12 weeks, more than 80% in both groups reported recurrent vulval itch. Conclusion: No significant differences in effectiveness were found between the two groups, because both treatments were ineffective. We can conclude that triamcinolone cream is therefore not the treatment of first choice for non-specific pruritus vulvae.

• Pruritus vulvae
• vulval disease
• topical steroids
• general practice
• RCT

Introduction

Pruritus vulvae is an annoying and shameful problem in women. A study in the United Kingdom reported that general practitioners (GPs) saw more than one patient a month with vulval symptoms [1]. It is known that recurrent vulval complaints may affect the psychological and sexual wellbeing of women. They report higher scores on depression scales and experience more psychological distress and reduced sexual satisfaction [2]. Pruritus vulvae is the most common symptom among vulvar complaints. The aetiology of pruritus vulvae is diverse [3]. Infection is the most prevalent cause in general practice, predominantly caused by Candida species. Some other specific but non-infectious causes of pruritus vulvae are postmenopausal mucosal atrophy, lichen sclerosus et atrophicus, and psoriasis [4], [6]. There is, however, a category of patients that suffer from pruritus vulvae without the presence of an infection or other specific causes. This subgroup of vulval complaints has rarely been studied and is therefore the subject of our research [7].

A pilot study performed in 25 Dutch general practices registering vulval complaints during 4 weeks showed approximately two new cases of pruritus vulvae without a specific cause per 1000 patients per year (unpublished data). This is in agreement with the aforementioned study in the United Kingdom [1]. Women with pruritus vulvae in this Dutch study were predominantly treated with topical steroids. Dutch GPs preferred topical steroids as the treatment of choice. Dutch GPs also mentioned aspects of patient education, such as avoiding cleaning and not using irritative materials as inlays. Nevertheless, the majority prescribed a topical steroid. The effectiveness of topical steroids is mainly studied in trials concerning the treatment of lichen sclerosus [8], [9]. Application of topical steroids in the vulvar region has long-term side effects, the most important being atrophy of the mucosa [10]. Because of these side effects, it is important to establish the benefits of topical steroids compared to less harmful alternatives. The aim of the study was to assess the effectiveness of treatment of non-specific pruritus vulvae with a topical steroid compared to a placebo.

Methods

The trial was performed in 25 general practices in the Netherlands, of which 13 were located in big cities and 12 in rural areas. The participating GPs were all visited and instructed by the researcher SS. All women between the age of 18 and 55 years presenting with acute as well as recurrent pruritus vulvae were informed about the study and consented to participate (informed consent). Pregnant and breastfeeding women and those who were unable to fill in the questionnaires were excluded. Women were also excluded when the pruritus lasted longer than 4 weeks or was caused by an infection or other specific cause such as postmenopausal atrophy, ulcer, malignancy, psoriasis, atopic dermatitis elsewhere, or lichen sclerosus. A cervical swab was performed to diagnose infections such as Candida albicans, Chlamydia trachomatis, Trichomonas vaginalis, or Neisseria gonorrhoeae[8]. While waiting for the results, the patient was asked to complete a questionnaire and return after 1 week to discuss the results. No treatment was permitted during this week. During the second appointment, the patient was either excluded (in the case of a positive swab) or included and randomized (in the case of a negative swab). The GP completed a questionnaire, in which the patient's complaints and findings of inspection of the vulvar region were specified. Also, the patient's characteristics such as age, parity, marital status, use of contraceptives, personal-hygiene practices, self-treatment practices, comorbidity, medication, and insurance status (private or state) were collected.

Randomization occurred in blocks of four. This was based on the expectation that the participating GPs would include four patients during the study period. All practices received four similar-looking tubes, numbered 1 to 4. Two contained a topical steroid (triamcinolone acetonide 0.1%) and two a placebo cream (cetomacrogol cream, consisting of isopropyl myristate, paraffin, and vaseline). The study was double blind; only one researcher (S.S.) knew which numbers corresponded with the steroid and which with the placebo. The sequence was different for each of the 25 practices. The participating pharmacist made sure that the two ointments were similar in colour, consistency, and smell. The choice of triamcinolone cream was based on a pilot study preceding this trial, which revealed that this class 2 steroid was the GPs’ treatment of choice for non-specific pruritus vulvae.

The patients received a tube with ointment and a checklist diary to score the severity of their itch daily. This diary was also used to monitor user compliance. All patients received a short list of behavioural advice.

To measure the severity of itch, we used an 11-point visual analogue scale, with a minimum score of zero (“no itch at all”) and a maximum of 10 (“the worst itch imaginable”). COOP/WONCA charts were used to assess changes in general health before and after treatment [12]. This instrument, used by GPs, was designed by the Dartmouth Primary Care Cooperative and WONCA (World Organization of Family Doctors), and comprises seven questions concerning health. In the COOP/WONCA questionnaire, possible answers were numbered from 1 to 5, in which 1 is a good state of function or health and 5 is a poor state of function or health. A questionnaire completed at the beginning of treatment registered patient characteristics and possible contributing factors to vulvar complaints, such as the use of hygienic materials, skin diseases elsewhere, comorbidity, and medication [13].

Patients were instructed by their GP to apply the ointment twice a day during 3 weeks and then return for a final consultation. At this stage, GPs were unaware of the kind of treatment the patient had received. GPs collected all registration forms and returned them to the researcher SS. Nine weeks after the end of treatment, patients were contacted by telephone by the research assistant to inquire about the presence of (relapse) complaints.

“Intention to treat” analysis was performed. We defined three outcome measures. First, mean changes in the itch VAS score were measured before and after treatment. Second, we calculated the number of women who were completely cured after treatment (VAS score zero). Change in general health perception measured by the COOP/WONCA charts was the third outcome measure.

We used SPSS version 12.0.1 for Windows to register all data. Statistical analysis was performed by using the chi-square test (significance p<0.05).

The study was approved by the ethics committee of the Radboud University Nijmegen Medical Centre, region Arnhem-Nijmegen (CWOM no. 0007-0157).

Results

The participating GPs enrolled 81 women presenting with pruritus vulvae (Figure 1). Sixteen patients were excluded because an infection was diagnosed, and 15 because, after 1 week, the complaints had subsided or they had refused to participate or did not show up. Fifty patients were randomized: 27 into the triamcinolone group and 23 into the placebo group. There was one drop out because too many data were missing. No significant differences were found between the two groups in patient characteristics such as age, marital status, parity, use of contraceptives, and insurance status (Table I). No differences were found in possible contributing factors to pruritus vulvae such as adverse personal hygiene, self-treatment, and comorbidity (data not shown). The mean severity of itch prior to the intervention was comparable for both groups. Also, compliance, as measured in patient diaries, did not show differences between the two groups.

Figure 1.  Flow chart of the study population.

Table I.  Patient characteristics of the study population.

	Triamcinolone group (n=26)	Placebo group (n=23)
Mean age, years (SD)	37.8 (11.9)	38.8 (9.3)
Marital status
Married/living together	19	15
Living with parents	–—	–—
Living alone	6	8
Mean parity (SD)	1.35 (1.33)	1.27 (1.35)*
Contraceptives	10	9
Oral contraceptives	4	7
Condom	3	1
Hormone, IUD	2	1
Insurance
State insurance	17	15
Private insurance	9	8
Itch level before treatment
Mean VAS score (SD)	5.60 (2.06)	6.48 (1.78)

* n=22; one person did not answer the question.

IUD: intrauterine device.

No significant differences.

The mean decrease in severity of itch after treatment was 2.08 points for the triamcinolone group and 3.26 points for the placebo group (non-significant, p=0.665) (Table II).

Table II.  Effect of treatment: changes in mean itch, as defined by VAS score.

Changes in mean itch	Triamcinolone group (n=26)	Placebo group (n=23)
Mean (SD)	−2.08 (−2.48)	−3.26 (−2.80)

The percentage of women that reached complete recovery was 42.3% in the triamcinolone group and 34.8% for the placebo group (non-significant, p=0.770). There was no difference in general health perception in the two groups, as measured with the COOP/WONCA charts.

Eight patients reported side effects: four in each group. In the triamcinolone group, one patient reported both “worsening of existing itch” and “other side effects”, and three patients simply reported “other side effects”. One patient in the placebo group experienced “worsening of the existing itch”; one reported “worsening of the itch” and “other side effects”; and two had “other side effects”.

Side effects described as “other” consisted of a burning sensation, vaginal discharge, abdominal pain, and low back pain for the placebo group. In the triamcinolone group, pain and a nasty feeling, allergic reaction, burning sensation, and itch around the anus were mentioned as “other side effects”.

The design of our study was double blind. Neither the treated patient nor the treating doctor was aware of the kind of treatment that was used. We checked the blinding after 3 weeks of treatment in both patient and doctor questionnaires. All indicated that they were still unaware of the kind of treatment they received (patients) or had given (doctors).

After 12 weeks, more than 80% of patients in both groups reported a relapse or continuation of pruritus vulvae.

Discussion

In this randomized placebo-controlled trial, we investigated the effectiveness of treating non-specific pruritus vulvae with a topical steroid in comparison with a placebo. After 3 weeks of treatment, we found no significant differences. In both groups, there was a reported diminishment of itch after 3 weeks, while less than half of the women completely recovered. In both groups, the relapse rate of vulvar itch after 12 weeks was high (80%). Based on our results, we conclude that treatment of non-specific pruritus vulvae with a topical steroid is not more effective than treatment with a placebo cream. Moreover, both treatments are ineffective in the long run. The reported side effects in both groups were comparable; one out of six patients experienced side effects, such as worsening of the itch or a burning sensation in the genital region. We expect side effects of topical steroids to increase when used over a longer period of time, especially atrophy [10]. Since non-specific pruritus vulvae is likely to be recurrent, women will probably use the ointment for a period much longer than 3 weeks and are therefore at risk of experiencing side effects at a later date.

Randomized controlled trials concerning the treatment of pruritus vulvae are rare. We did not find any studies carried out in a general healthcare setting. We therefore believe our study contributes to present knowledge of the treatment of this condition. The size of our study sample does, however, limit our conclusions. Small differences in effectiveness may not have been detected. However, this would raise the question as to what the clinical relevance of these small differences would be. We believe that only distinct differences in effectiveness should influence the choice of treatment and can justify the use of an ointment that has potential long-term side effects.

An infection is the most common cause of vulval itch [3]. Therefore, a cervical swab was performed preceding the start of treatment. We expected a large number of positive swabs and thus a large number of exclusions, of at least 50% (14). Surprisingly, only 28% of the women presenting with pruritus vulvae were excluded because of a positive swab. Since the participating GPs were well aware of the goal of this study, we presume that they may have pre-selected women that might fit the study design. However, since this remains a matter of assumption, it is hard to estimate the effect of this pre-selection on our results. We expect this phenomenon to have affected our study group as a whole, without causing a bias between the two groups. Women were motivated to take part. We had only one drop out due to missing questionnaires. All women completed the treatment. The randomization was therefore sufficiently adequate.

Conclusion

Non-specific pruritus vulvae is a harmless but nevertheless annoying complaint which can be shameful and embarrassing. Adverse personal-hygiene practices such as the daily use of panty liners, deodorants, excessive washing of the vulva, or self-medication must be avoided.

Triamcinolone cream lacks clear evidence of effectiveness. Therefore, women should no longer be treated with topical steroids as part of standard care. Women should also be informed that vulvar itch is very likely to relapse and that an ordinary cream such as cetomacrogol is the treatment of choice, without long-term side effects.

Acknowledgements

We would like to thank Frank Martens and Hans Bor for help with statistics. The present study was funded by a grant from the Netherlands Organisation for Health Research and Development (ZonMw) and the Dutch Association of General Practitioners (NHG). Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.