From the Institute of General Practice, University of Aarhus, Aarhus, Denmark. Requests for reprints should be addressed to Flemming Bro, MD, PhD, Institute of General Practice, University of Aarhus, Finsensgade 10, DK-8000 Aarhus C, Denmark.
Vaginal discharge and pruritus are frequent reasons for encounter in general practice, [1] and Candida vaginitis is the most frequent diagnosis made in these patients.[2] Numerous studies have evaluated different treatment regimens for vaginal candidiasis, and boric acid, nystatin, natamycin, and a range of different imidazole derivatives have all proved effective in the eradication of Candida.[3-6] Most clinical trials have compared different drugs or different dosages of the same drug. In a review, Odds[7] concluded that imidazole derivatives gave higher cure rates than polyenes (eg, nystatin), but in the individual studies no statistical difference was usually found. Placebo-controlled studies are infrequent, and only little is known about the natural history of Candida vaginitis.
Since imidazole was introduced in the early 1970s, local treatment with imidazole derivatives has been the first choice for vaginal candidiasis. The duration of treatment has gradually been reduced from 1 week to 3 days without a decline in the cure rates,[5,8,9] and single-dose treatment with vaginal tablets is now recommended in uncomplicated cases.[10,11]
Most studies on Candida vaginitis have been conducted in hospitals or venereal disease clinics. Women seen in general practice differ from such populations, however, [12] and results obtained in those studies cannot be applied to a general practice population without consideration.
The aim of the present study was to compare single-dose 500-mg clotrimazole vaginal tablets with placebo in the treatment of Candida vaginitis in general practice.
METHODS
This multipractice study included 29 general practitioners in the county of Aarhus, Denmark. The study was approved by the county's ethics committee. Each general practitioner was asked to enroll 15 consecutive patients with increased vaginal discharge, pruritus, or malodor. After giving their informed consent, the women underwent a standardized procedure, including history and pelvic examination. Patients were excluded for the following reasons: age less than 18 years, pregnancy, menstruation or expected menstruation within the next 8 days, and treatment with antibiotics within the previous 14 days.
A total of 361 women with a mean age of 28.7 years (SD 8.8 years) was included in the study. All the women, except 129 with bacterial vaginosis according to the definition of Amsel et al,[13] were randomly assigned to treatment with single-dose 500-mg clotrimazole vaginal tablets or placebo. The analysis excluded 95 women in whom no Candida was found by either of the two culture methods, as well as 38 in whom Neisseria gonorrhoeae, Chlamydia trachomatis, or Trichomonas vaginalis were isolated. The treatment trial hereafter comprised 99 women with Candida found by at least one of the two culture methods. In one patient randomized to treatment with clotrimazole and in three patients treated with placebo, information about culture result or symptoms after treatment was missing, and subsequent analysis comprised 55 women randomized to treatment with clotrimazole and 40 to placebo.
Seven to 10 days after the first visit, a follow-up visit was performed following the same format as at the first visit, including culture for Candida, and 4 weeks later a questionnaire was sent to all the women asking about the presence of vaginal symptoms. The study was conducted on a double-blind basis. The history was obtained and clinical examination and microbiological investigations were performed without knowledge of the treatment given.
Clinical Observation and investigation
During the pelvic examination, the appearance of the vaginal discharge was noted, and its pH was measured with a paper indicator (Merck, narrow range 4.0 to 7.0). A 10% potassium hydroxide (KOH) solution was then added to the discharge left on the withdrawn speculum, and the presence of a fishy odor was registered (amine test). A wet mount was examined by bright field microscopy at x 400 magnification for the presence of T vaginalis and clue cells (vaginal epithelial cells studded with coccobacilli). A diagnosis of bacterial vaginosis was made if at least three of the following four criteria were present: (1) a homogeneous vaginal discharge of thin appearance, (2) a pH [greater than] 4.5, (3) a fishy amine odor on addition of 10% KOH to the discharge, and (4) the presence of clue cells.[13]
Microbiological Methods
Cervical specimens for the detection of C trachomatis were obtained with cotton-tipped swabs, transported in a sucrose-phosphate medium to the Institute of Medical Microbiology, University of Aarhus, and cultured on cycloheximide-treated McCoy cells.[14]
Specimens for culture for N gonorrhoeae were obtained by charcoal cotton-tipped swabs from the cervix, and transported in Stuart's medium to the State Serum Institute (SSI), Copenhagen. N gonorrhoeae was isolated and identified according to standard laboratory procedures.[15]
Specimens for culture of T vaginalis and Candida were obtained by charcoal cotton-tipped swabs from the posterior fornix of the vagina and transported in Stuart's medium to SSI, Copenhagen. The swab for detection of T vaginalis was incubated in Diamond's medium, and isolation and identification of T vaginalis were performed according to procedures previously described.[16]
The swab for detection of Candida was plated onto Sabouraud maltose agar and incubated at 36[degrees]C for 3 days. Individual morphologically different colonies were streaked on cornmeal agar and incubated for 3 days at 25[degrees]C.[17] The plates were examined daily for the occurrence of chlamydospores, differentiating Candida albicans from other yeast species. The term Candida in this report includes all yeasts found.
One swab with vaginal secretion from the posterior fornix of the vagina was used for culture of Candida in the general practitioner's laboratory. The swab was inoculated in a liquid medium containing trypticase to which sheep serum, penicillin, and streptomycin were added. After incubation at 37[degrees]C for 24 hours, Candida was demonstrated by microscopy at x400 magnification. Species identification was not possible.
Statistics
Statistical analyses were performed with Pearson's chi-square test to compare the following in the two treatment groups: the characteristics of patients, the presence of Candida after treatment, the patients' evaluation of symptoms after treatment, and the patients' evaluation of symptoms in relation to the presence of Candida after treatment. In the evaluation of symptoms after treatment, power calculation was performed to determine the type II error (beta) according to Feinstein.[18]
RESULTS
Candida was isolated in the general practitioner's laboratory in 74 patients and at the bacteriological laboratory in 83 patients and by one or both methods in 95 women.
The characteristics of the patients are shown in Table 1. All characteristics were comparable in the two treatment groups (P [greater than] .05).
At the follow-up visit, 34 (62%) women treated with clotrimazole were culture negative for Candida, compared with 10 (25%) in the placebo group (95% confidence interval [CI], + 18% to +55%).
No difference in the mycological cure rate was found when patients with Candida isolated in the general practitioner's laboratory and at the bacteriological laboratory were analyzed separately.
If the analysis was restricted to women harboring the C albicans species, 67% (27/41) were culture negative after treatment with clotrimazole, compared with 19% (4/21) given placebo (95% CI, +24% to +68%).
Symptoms had improved or disappeared in 69% treated with clotrimazole, compared with 55% in the placebo group (95% CI, -6% to +34%). Assuming a minimal relevant difference of 30%, the probability of type II error is 5.6%.
Although symptoms 1 week after treatment were more frequently present in women who were still Candida positive, 23% of the Candida-negative patients still had symptoms, while 51% of the culture-positive patients stated that their symptoms had improved or disappeared.
The follow-up questionnaires sent I month after treatment were returned by 62 (65%) of the women with complete information. In both treatment groups 50% of the women reported that their vaginal symptoms were unchanged or worse (95% CI, -26% to +26%). Assuming a minimal relevant difference of 30%, the probability of type II error is 1.2%.
DISCUSSION
In previous studies of single-dose 500-mg clotrimazole in the treatment of Candida vaginitis, Candida was eliminated in 74% to 94%.[5,9,10,19] None of the treatment trials was from general practice, and the lower mycologic cure rate in the present study may be due to differences between the study populations, eg, with regard to the virulence and antibiotic resistance of the Candida, differences in sexual transmission, gastrointestinal colonization, and perhaps dietary habits.
The special interests of investigators carrying out a hospital-based study are likely to improve compliance and reduce the number of inappropriate applications of the vaginal tablets. In some studies application was done by the investigator, thereby eliminating any problems with compliance. The present study was carried out by a number of general practitioners under conditions very similar to everyday routine. Low compliance seems less likely in this study because single-dose therapy was used, but inappropriate application of the vaginal tablets could explain the lower mycologic cure rate. Results obtained under these circumstances do, however, reflect reality in general practice rather than what could be obtained under optimal conditions.
From a pharmacologic and microbiologic point of view, the primary goal in the treatment of vaginal candidiasis is to eradicate Candida. From a clinical point of view the objective is to relieve symptoms, since the presence of Candida is in itself not an indication for therapy.
More than 83% of the patients were free of symptoms 1 week after treatment in previous studies using single-dose clotrimazole 500 Mg.[5,9] With the lower mycologic cure rate in the present trial, one would also expect a lower symptomatic cure rate. The use of placebo, however, may also tend to reduce the symptomatic cure rate because the patients' expectations of the outcome of the treatment may have been lower, since they knew that they might receive placebo.
Active treatment did not significantly improve the symptomatic cure rate compared with placebo 1 week after treatment, and no therapeutic benefit was found after 5 weeks. Unfortunately the large number of women lost to follow-up introduces a bias. It is difficult to evaluate how those responding to the questionnaires differ from nonresponders, but the dropout rate reduces the validity of the long-term evaluation.
Candida is frequently present in women without vaginal symptoms, and it is therefore considered to be an opportunistic pathogenic microorganism.[20,21] Candida would also be expected to be present in a proportion of women with vaginal complaints without being the cause of symptoms. In these women the eradication of Candida would not affect the symptoms, which might explain that 23% of the women in whom no Candida was found after treatment still had symptoms.
Though troublesome, vaginal candidiasis is not a serious condition, and none of the women left the study because of intolerable symptoms before the first follow-up after 1 week. Symptoms improved or disappeared in more than one half of the women on placebo after 1 week, and active treatment was not significantly better than placebo. Further clinical trials conducted in general practice and including placebo therapy are needed to confirm these results and to evaluate other regimens in the treatment of Candida vaginitis.
Acknowledgments
This study was supported by grants from The Danish Medical Research Council. Medication was provided by Bayer Danmark A/S.
* TABULAR DATA OMITTED
References
1. Fry J: Common Diseases, Their Nature, Incidence and Care, ed 2.
Lancaster, England, Medical & Technical Publishing, 1979
2. Bro F: The diagnosis and treatment of patients with vaginal discharge
in general practice. Ugeskr Laeger 1988; 150:1553-1556
(English summary)
3. van Slyke KK, Michel VP, Rein MF: Treatment of vulvovaginal
candidiasis with boric acid powder. Am J Obstet Gynecol 1981;
141:145-149
4. Eliot BEW, Howat RCL, Mack AE: A comparison between the effects
of nystatin, clotrimazole and miconazole on vaginal candidiasis. Br J
Obstet Gynaecol 1979; 86:572-577
5. Robertson WH: A concentrated therapeutic regimen for vulvovaginal
candidiasis. JAMA 1980; 244:2549-2550
6. Buch A, Christensen ES: Treatment of vaginal candidosis with natamycin
and effect of treating the partner at the same time. Acta
Obstet Gynecol Scand 1982; 61:393-396
7. Odds FC: Candida and Candidosis. London, Leicester University
Press, 1979
8. Highton BK: A trial of clotrimazole and nystatin in vaginal moniliasis.
J Obstet Gynecol 1973; 80:992-995
9. Milsom I, Forssman L: Treatment of vaginal candidosis with a single
500-mg clotrimazole pessary. Br J Vener Dis 1982; 58:124-126
10.Cohen L: is more than one application of an antifungal necessary in
the treatment of acute vaginal candidiasis? Am J Obstet Gynecol
1985; 152:961-964
11.Fleury F, Hughes D, Floyd R: Therapeutic results obtained in vaginal
mycoses after single dose treatment with 500 mg clotrimazole vaginal
tablets. Am J Obstet Gynecol 1985; 152:968-970
12.Bro F: Patients with vaginal discharge in general practice. Acta
Obstet Gynecol Scand 1989; 68:41-43
13.Amsel R, Totten PA, Spiegel CA, et al: Nonspecific vaginitis. Am J
Med 1983; 74:14-23
14.Ripa KT, Mardh P-A: Cultivation of Chlamydia trachomatis in cycloheximide-treated
McCoy cells. J Clin Microbiol 1977; 6:328-331
15.Reyn A: Recent developments in the laboratory diagnosis of gonococcal
infections. Bull WHO 1969; 40:245-255
16.Nielsen R: Trichomonas vaginalis. Laboratory investigations in trichomoniasis.
Br J Vener Dis 1973; 49:531-535
17.McGinnis MR: Laboratory Handbook of Medical Mycology. New
York, Academic Press, 1980
18.Feinstein AR: Clinical biostatistics-XXXIV. The other side of "statistical
significance": alpha, beta, delta, and the calculation of sample
size. Clin Pharmacol Ther 1975; 18:491-W5
19.Goormans E, Bergstein NAM, Loendersloot EW, Branolte JH: One-dose therapy of
Candida vaginitis. Chemotherapy 1982; 28:106-109
20.Sobel JD: Epidemiology and pathogenesis of recurrent vulvovaginal
candidiasis. Am J Obstet Gynecol 1985; 152:924-934
21.Gardner HL: Candidiasis. in Gardner HL, Kaufman RH (eds): Benign
Diseases of the Vulva and Vagina. Boston, GK Hall Medical, 1981,
pp 217-242
COPYRIGHT 1990 Dowden Health Media, Inc.
COPYRIGHT 2004 Gale Group