Prevalence of Candida Glabrata and Its Response to Boric Acid Vaginal

Pathophysiology/Complications ORIGINAL ARTICLE

Prevalence of Candida glabrata and Its Response to Boric Acid Vaginal Suppositories in Comparison With Oral Fluconazole in Patients With Diabetes and Vulvovaginal Candidiasis

1 4 DEBARTI RAY, MBBS DEEPTI GOSWAMI, MD, MRCOG brata VVC respond poorly to single oral 1 2 RAVINDER GOSWAMI, MD, DM PIYALI MANDAL, MSC doses of 150 mg fluconazole (8). C. gla- 2 5 UMA BANERJEE, MD VISHNUBHATLA SREENIVAS, PHD 3 1 brata demonstrates intrinsically reduced VATSLA DADHWAL, MD NARAYANA KOCHUPILLAI, MD susceptibility to azole drugs (10–13). Redondo-Lopez et al. (10) and other investigators have reported the successful OBJECTIVE — A large proportion of vulvovaginal candidiasis (VVC) in diabetes is due to use of vaginal boric acid suppositories in non–albicans Candida species such as C. glabrata and C. tropicalis. Observational studies indicate the management of patients with C. gla- that diabetic patients with C. glabrata VVC respond poorly to azole drugs. We evaluated the brata (10,14,15) as well as C. albicans response to oral fluconazole and boric acid vaginal suppositories in diabetic patients with VVC. VVC (16–18). In view of the higher frequency of C. glabrata VVC among diabetic RESEARCH DESIGN AND METHODS — A total of 112 consecutive diabetic patients with VVC were block randomized to receive either single-dose oral 150-mg fluconazole or boric women and their poor response to flucon- acid vaginal suppositories (600 mg/day for 14 days). The primary efficacy outcome was the azole, there is a need for a therapeutic reg- mycological cure in patients with C. glabrata VVC in the two treatment arms. The secondary imen that is effective against both C. outcomes were the mycological cure in C. albicans VVC, overall mycological cure irrespective of glabrata and C. albicans. To the best of our the type of Candida species, frequencies of yeast on direct microscopy, and clinical symptoms knowledge, there is no previous study re- and signs of VVC on the 15th day of treatment. Intention-to-treat (ITT; n ϭ 111) and per- porting the effect of boric acid vaginal protocol (PP; n ϭ 99) analyses were performed. suppositories in diabetic women with VVC. In the current open-label, random- RESULTS — C. glabrata was isolated in 68 (61.3%) and C. albicans in 32 (28.8%) of 111 ized trial, we report the response to oral subjects. Patients with C. glabrata VVC showed higher mycological cure with boric acid com- fluconazole and boric acid vaginal sup- pared with fluconazole in the ITT (21 of 33, 63.6% vs. 10 of 35, 28.6%; P ϭ 0.01) and PP analyses (21 of 29, 72.4% vs. 10 of 30, 33.3%; P ϭ 0.01). The secondary efficacy outcomes were positories in diabetic patients with VVC. not significantly different in the two treatment arms in the ITT and PP analyses. RESEARCH DESIGN AND CONCLUSIONS — Diabetic women with C. glabrata VVC show higher mycological cure with boric acid vaginal suppositories given for 14 days in comparison with single-dose oral METHODS — This was a random- 150-mg fluconazole. ized, open-label comparison of 2 weeks’ duration between single-dose oral Diabetes Care 30:312–317, 2007 150-mg fluconazole and 14 days of treatment with boric acid vaginal supposito- atients with diabetes are at increased infection due to non–albicans Candida ries (600 mg/day) in 112 consecutive patients with diabetes (type 2 diabetes, risk of developing vulvovaginal can- species such as C. glabrata and C. tropica- ϭ ϭ P didiasis (VVC) (1–8). Unlike nondi- lis (4,6–8). In contrast, C. albicans VVC is n 77 and type 1 diabetes, n 35) and abetic women, these patients have a more frequent in nondiabetic women (9). VVC conducted during 2004–2006. Dia- higher proportion of colonization/ Moreover, diabetic patients with C. gla- betes was diagnosed as per the American Diabetes Association Expert Committee ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● criteria (19). Patients already on insulin or From the 1Department of Endocrinology and Metabolism, All India Institute of Medical Sciences, New Delhi, 2 oral hypoglycemic agents also were in- India; the Department of Microbiology, All India Institute of Medical Sciences, New Delhi, India; the cluded. Any patient with onset of diabetes 3Department of Obstetrics and Gynecology, All India Institute of Medical Sciences, New Delhi, India; the 4Department of Obstetrics and Gynecology, Maulana Azad Medical College, New Delhi, India; and the before age 30 years who had received con- 5Department of Biostatistics, All India Institute of Medical Sciences, New Delhi, India. tinuous insulin treatment since diagnosis Address correspondence and reprint requests to Dr. Ravinder Goswami, MD, DM, Associate Professor, was considered to have type 1 diabetes. Department of Endocrinology and Metabolism, All India Institute of Medical Sciences, New Delhi 110029, Subjects excluded were those who were India. E-mail: Ravinder Goswami at [email protected] or Uma Banerjee at uma_banerjee@hotmail. pregnant, sexually inactive girls, those com. Ͼ Received for publication 18 July 2006 and accepted in revised form 22 October 2006. aged 65 years, those with renal insuffi- Abbreviations: HVS, high vaginal swab; ITT, intention to treat; PP, per-protocol; VVC, vulvovaginal ciency (serum creatinine Ͼ1.8 mg/dl), candidiasis. and those on steroid therapy. Patients DOI: 10.2337/dc06-1469. Clinical trial reg. no. NCT00353561, clinicaltrials.gov. who did not give consent for pelvic exam- © 2007 by the American Diabetes Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby ination, those treated for vaginal dis- marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. charge during the past 3 months, and

312 DIABETES CARE, VOLUME 30, NUMBER 2, FEBRUARY 2007 Ray and Associates symptomatic patients in whom Candida number of patients in the two treatment 150 mg fluconazole. Mycological cure growth was not detected on fungal culture arms. was defined as the absence of Candida were also excluded. A1C was measured in A statistician generated the entire ran- growth on the HVS culture on the 15th all the subjects to assess their glycemic domization sequence list in advance, and day of therapy. The primary efficacy out- control (normal range: 5.4–7.0%) at visit allocations were sequentially numbered come was the mycological cure in patients 1. The institutional ethics committee of from the beginning to end. A female clin- with C. glabrata VVC in the two treatment the All India Institute of Medical Sciences ical investigator (D.R.) carried out genital arms. The secondary efficacy outcomes approved the study protocol, and all the examinations for VVC, including an HVS were mycological cure in C. albicans VVC, subjects were examined after their infor fungal culture for all consecutive dia- overall mycological cure irrespective of formed consent. betic patients attending the endocrine the type of Candida species isolated, pres- outpatients clinic and gave a serial num- ence of yeast on direct microscopy, and ber to all the patients examined. Follow- frequency of clinical symptoms and signs Diagnosis of VVC ing a report of yeast growth, two of VVC on the 15th day in the two treat- A large number of diabetic patients were clinicians (D.R. and R.G.) assigned pa- ment arms. assessed for symptoms and signs of VVC tients to a treatment arm, as per the serial including vaginal discharge, vulval pruri- order of the clinical examination and ran- Adverse effects tus, burning sensation, vulval edema, and domization list sequence number, in an Treatment-emergent adverse events were vaginal congestion at visit 1 (20). Two irreversible manner. One of these two cli- defined as those occurring after receiving sterile cotton-tipped commercial swabs nicians (R.G.) was not involved with clin- the first dose of treatment. were used to collect discharge from high ical examination. Patients were allocated vagina and transported to the mycology to treatment arm following reports of Sample size and power calculations laboratory without delay. One of the yeast growth but before the results of Can- Sample size estimation for the assessment swabs was used to determine the presence dida species were available. The serial or- of primary efficacy outcome was based on of yeast by direct microscopy, while the der of the randomization list was not our previously published mycological other was used for fungal culture. The disclosed to the patients until the treat- cure rate of 18.7% in diabetic patients VVC was diagnosed in the presence of ment was assigned. with C. glabrata VVC after a single oral clinical signs and symptoms and growth The intake of fluconazole was super- dose of 150 mg fluconazole therapy (8). of Candida species on culture of the high vised to ensure compliance. Patients re- The sample size was calculated using a vaginal swab (HVS). ceiving boric acid were instructed to statistical program (EPI-Info 2002; Cen- Detection and identification of the insert one boric acid–filled gelatin capsule ters for Disease Control and Prevention, yeast up to the species level was done in (600 mg) intravaginally every night for 14 Atlanta, GA), assuming the persistence of the mycology laboratory of the All India uninterrupted days either by a plastic ap- the above trend and three times higher Institute of Medical Sciences as per the plicator or dispensed by hand as per their response rate (56.1%) with boric acid standard protocols (21,22). The criteria preference. Two extra capsules were therapy. Calculations revealed that 30 for laboratory diagnosis were 1) direct given, and unused capsules were counted subjects with C. glabrata infection in each demonstration of the yeast by Gram’s on the follow-up visit to assess compli- of the two treatment arms would be re- staining and 2) culture on Sabouraud’s ance. Patients in both treatment arms quired to give a study power of 80% at dextrose agar supplemented with 2 were reassessed on the 15th day of ther- 95% CI. With a 54.1% prevalence rate of mg/ml gentamicin, with and without cy- apy (visit 3). To minimize interobserver C. glabrata infection, as observed in our clohexamide (0.5%). Identification of variation, the same investigator carried previous study in diabetic subjects with Candida species was performed by exam- out the repeat clinical assessment at visit 3 VVC (8), a sample size of 112 diabetic ining colony morphology, germ tube test, when HVSs were taken again for demon- patients with VVC was estimated to give hyphal morphology, and chlamydospore stration of yeast by direct microscopy and us 60 subjects with C. glabrata VVC. formation on cornmeal agar, triphenyl fungal culture. All the HVSs were of the tetrazolium reduction test, fermentation same color and size and sent to the myco- Statistical analysis and assimilation of different sugars, and logical laboratory in identical containers Both intention-to-treat (ITT) and per- cycloheximide susceptibility. labeled with a numerical identity. The protocol (PP) analyses were performed as plan of block randomization, patients’ efficacy analyses. Patients included in the Randomization and concealment treatment assignments, the pre- and post- ITT analysis included all those who were Following identification of Candida nature of the specimen and clinical find- assigned to the two treatment arms and in growth on HVSs, subjects were assigned ings were not disclosed to the mycologist whom at least baseline HVS culture to receive single-dose oral 150-mg flu- and the laboratory staff who performed showed Candida growth. To be included conazole or boric acid vaginal supposito- fungal culture and species identification in the PP analysis, patients also had to ries (600 mg/day) for 14 days in a 1:1 ratio till the completion of the 2-year study. comply with the assigned treatment regi- using a randomization list with random men and follow-up as per the study permutated blocks, length of 4, at the clinic Efficacy outcomes protocol on day 15 (visit 3). For determi- site (visit 2). The randomization list was The primary objective was to test the hy- nation of the primary and secondary effi- generated using a Microsoft Excel spread- pothesis that diabetic patients with C. gla- cacy outcomes in the ITT analysis, sheet as described in detail (available at brata VVC when treated with boric acid missing data were imputed by explicit al- http://www.childrens-mercy.org/stats/ vaginal suppositories (600 mg/day) for 14 location of poor outcome in both treat- plan/random.asp). Block randomization days show a higher mycological cure rate ment arms. The differences in the mean was used to ensure the balance in the in comparison with a single oral dose of age, BMI, and A1C between patients in

DIABETES CARE, VOLUME 30, NUMBER 2, FEBRUARY 2007 313 Diabetes and vulvovaginal candidiasis

Table 1—Baseline clinical characteristics, symptoms, and signs and type of Candida species in the ITT and PP groups

ITT analysis PP analysis Parameters Fluconazole Boric acidP value Fluconazole Boric acid P value n 55 56 49 50 Age (years) 40.2 Ϯ 10.7 41.2 Ϯ 11.3 0.63 40.2 Ϯ 9.8 40.8 Ϯ 11.5 0.78 BMI (kg/m2) 24.4 Ϯ 5.4 24.9 Ϯ 4.5 0.64 24.6 Ϯ 5.6 24.7 Ϯ 4.7 0.88 A1C (%) 9.36 Ϯ 2.5 8.5 Ϯ 1.6 0.04 9.35 Ϯ 2.8 8.5 Ϯ 1.7 0.05 Postmenopausal 21 (38.1) 17 (30.3) 0.50 19 (38.8) 14 (28.0) 0.35 Use of commercial sanitary napkins during 14 (41.2) 22 (56.4) 0.28 13 (43.3) 22 (61.1) 0.23 menstruation VVC symptoms for Ͼ1 month 50 (90.9) 47 (83.9) 0.41 44 (89.8) 41 (82.0) 0.40 Vulval pruritus 45 (81.8) 41 (73.2) 0.39 40 (81.6) 35 (70.0) 0.26 Vulval edema 5 (9.0) 13 (23.2) 0.08 5 (10.2) 9 (18.0) 0.40 Vaginal congestion 39 (70.9) 43 (76.7) 0.62 33 (67.3) 37 (74.0) 0.61 Direct microscopy positivity 35 (63.6) 30 (53.5) 0.37 29 (59.2) 24 (48.0) 0.36 C. glabrata 35 (63.6) 33 (58.9) 0.75 30 (61.2) 29 (58.0) 0.90 C. albicans 14 (25.4) 18 (32.1) 0.95 13 (26.5) 16 (32.0) 0.92 Other Candida species 6 (10.9) 5 (8.9) 0.97 6 (12.2) 5 (10.0) 0.97 Data are means Ϯ SD for age, BMI, and A1C and n (%) for other indices in the two treatment arms in ITT and PP groups. P values test differences in the means or proportions observed in the two treatment arms. the two treatment arms were analyzed us- 28.8%) and C. tropicalis (n ϭ 4, 3.6%). analyzed irrespective of the type of Can- ing Student’s t test. Logistic regression The baseline demographic and clinical dida species isolated at baseline HVS cul- analysis, with mycological cure as the de- characteristics, including the proportion ture, tended to be higher in those pendent variable and treatment group of patients with type 1 and type 2 diabe- receiving boric acid in the ITT (P ϭ 0.07) and A1C as the independent variable, was tes, frequencies of C. glabrata and C. albi- and PP (P ϭ 0.06) analyses. The improve- used to determine the significance of dif- cans species isolated on HVS culture, ment in vaginal pruritus, discharge, and ference in the primary and secondary ef- number of postmenopausal women, and other clinical signs and symptoms was ficacy outcomes in the two treatment those using commercially available pro- comparable in the two treatment arms in groups. The differences in the parametric tective sanitary napkins, were comparable the ITT or PP analyses. and nonparametric indexes between sub- in both the treatment arms in the ITT and The duration of different symptoms jects who grew C. albicans and C. glabrata PP analyses groups (Table 1). Only two and signs related to VVC, yeast positivity on HVS culture were compared using a women were on oral contraceptive pills on direct microscopy, and mean A1C val- Student’s t test and ␹2 test, respectively. (one in each of the two treatment arm). ues were comparable in diabetic subjects All P values calculated were two tailed, The mean A1C value at baseline was who grew C. albicans and C. glabrata in and P Ͻ 0.05 was considered significant. lower in the boric acid treatment arm than the ITT and PP analyses groups. The SPSS 7.5.1 statistical package was used in the fluconozaole arm in the ITT as well mean A1C tended to be higher in subjects for analysis. as in the PP analyses group. The primary who continued to grow Candida on repeat and secondary efficacy outcomes were, HVS culture compared with those who RESULTS — Of 112 subjects random- therefore, adjusted for A1C at baseline did not grow Candida in the ITT (9.3 Ϯ ized to the two treatment arms, 111 were (Table 2). 2.0% vs. 8.6 Ϯ 2.1%; P ϭ 0.11) and PP included in the ITT analysis and 99 were The mycological cure rate in patients groups (9.5 Ϯ 2.2% vs. 8.6 Ϯ 2.1%; P ϭ analyzed in the PP analysis. One patient with C. glabrata VVC was significantly 0.06). was excluded from the ITT analysis be- higher in the boric acid treatment arm cause she fell into exclusion criteria (sex- compared with the fluconazole arm in Adverse effects ually inactive, in whom a HVS was not both the ITT and PP analyses (P values ϭ Two of the patients in the boric acid treat- taken). Twelve patients were dropped for 0.01 for both after adjustment for baseline ment arm who stopped the treatment due PP analysis because of the following rea- A1C). The odds in favor of mycological to vaginal burning sensation were consid- sons: two stopped boric acid therapy due cure with boric acid vaginal suppositories ered to demonstrate the adverse effect, as to vaginal burning sensation appearing on in diabetic patients with C. glabrata VVC this improved after stopping the drug. the 7th day of therapy, two did not come was 4.0 (95% CI 1.4–11.6). The good- for follow-up because of diabetic foot and ness of fit of the logistic regression model CONCLUSIONS — The 59.9% prev- pulmonary consolidation (fluconazole was assessed by Nagelkerke R2. Although alence of C. glabrata infection observed in arm), and eight patients were lost to fol- the Nagelkerke R2 value for primary out- the current study confirms the findings of low-up for unknown reasons (four in come was only 15.3%, the model should our earlier studies that non-albicans VVC each of the two treatment arms). not be interpreted as inadequate (23). No is frequent in diabetic women (4,8). de The most common species isolated in significant difference in mycological cure Leon et al. (6) observed the 54% vaginal the cohort of 111 patients included in the was observed in the two treatment arms in carriage rate of C. glabrata in type 2 dia- ITT analysis was C. glabrata (n ϭ 68, subjects with C. albicans VVC. The overall betes. The comparable frequency of clin- 61.3%), followed by C. albicans (n ϭ 32, mycological cure on the 15th day, when ical symptoms and signs between diabetic

314 DIABETES CARE, VOLUME 30, NUMBER 2, FEBRUARY 2007 Ray and Associates ucms(fe dutetfrACa aeie.Ptet ihmxdifcinadnon and infection mixed with Patients baseline). at A1C for adjustment (after outcomes aaaeod ai 9%C)or CI) (95% ratio odds are Data 2— Table eodr outcomes Secondary outcome Primary Parameters

uvleea6(02 1.)11(.–.)09 .0 20 A08 NA 0.89 NA 0.117 0.075 0.09 0.09 (2.0) 1 (0.02–1.4) 0.15 (0.2–1.1) women 0.48 0.053 0.096 (2.0) 1 (38.0) — 19 0.07 with 0.06 (0.13–1.1) C. 0.4 (57.1) (10.2) 28 5 (0.97–5.5) 0.191 glabrata 2.3 0.007 (14.0) 7 0.059 0.038 (74.0) 37 0.01 0.91 or C. 0.09 (1.4–14.4) (30.6) 4.5 15 0.20 (0.3–3.4) (51.0) 1.1 albicans 25 (0.2–1.1) 0.5 (0.2–1.4) above 0.5 As 0.034 (12.5) 7 0.069 in- (44.6) 25 (12.5) 7 0.10 above As 0.07 (10.2) 6 (61.8) 34 (0.2–1.2) 0.5 (20.0) (0.9–4.5) 11 0.153 2.1 (23.2) 13 (66.1) 37 0.01 (38.2) (1.4–11.6) 21 4.0 (45.4) congestion 25 Vaginal edema Vulval above As pruritus Vulval microscopy direct on Yeast cure mycological Overall above As in cure Mycological A1C) for (adjusted above As in cure Mycological

fection is similar to that reported by Gei- VVC VVC ger et al. (24) in nondiabetic women.

Increased prevalence of C. glabrata groups PP and ITT the in arms treatment two the in outcomes efﬁcacy secondary and Primary infection in diabetic women has clinical relevance because poor therapeutic response and innate resistance to azoles has been reported for C. glabrata VVC in non- C.albicans C.glabrata diabetic women (10–13). Similar infor- n

mation is lacking in diabetic subjects, as groups. PP and ITT in arms treatment two in (%) they are often excluded in antifungal efﬁ- cacy studies (8,10,25–27). Poor mycolog- 03 2.)2/3(36 . 161.)0050191/0(33 12 7.)53(.–60 .0 0.194 0.003 (1.7–16.0) 5.3 (72.4) 21/29 (33.3) 10/30 0.159 0.005 (1.6–12.1) 4.4 (63.6) 21/33 (28.6) 10/35 Fluconazole 2(57 1(11 . 00–.)01 .1 2(23 1(88 . 00–.)01 0.138 0.14 (0.02–1.8) 0.1 (68.8) 11 (92.3) 12 0.110 0.13 (0.04–1.5) 0.3 (61.1) 11 (85.7) 12 ical cure in diabetic women with C. ( n glabrata VVC to single-dose oral 150-mg ϭ

fluconazole, observed in the current 55) study, is in accordance with our earlier case-control study (8). Mechanisms im- plicated for resistance of C. glabrata to flu- oi acid Boric ( conazole include increased fungal n ergosterol synthesis and up to eightfold ϭ higher expression of azole efflux pump 56) protein coded by CgCDR1 transporter genes (26,28). Diabetic patients with analysis PP analysis ITT

VVC also respond poorly to itraconazole ratio Odds P 9%CI) (95% and ketoconazole (29,30) secondary and primary for groups PP and ITT the in arms treatment two the in observed proportions the in differences test values

In the current study, the higher my- -albicans cologoical cure (72.4%) to boric acid therapy in diabetics with C. glabrata VVC is similar to that reported in nondiabetic than other candidiasis individuals (10,14–18). In 1974, Swate P and Weed (16) ﬁrst reported boric acid to value be a “safe, effective and inexpensive form of therapy” for VVC. Sobel and Chaim kerke- (14) reported clinical improvement in Nagel-

81% and mycological cure in 77% of non- glabrata C. R diabetic patients with symptomatic C. gla- 2 brata/Torulopsis vaginitis treated with vaginal boric acid. There are also reports Fluconazole eeecue nteaayi eas ftelmtdnumbers. limited the of because analysis the in excluded were ( of better therapeutic response to boric n ϭ

acid compared with topical nystatin and 49) miconazole in nondiabetic women with C. albicans VVC (17,18). Otero et al. (31) have reported a greater in vitro suscepti- oi acid Boric (

bility of C. albicans to boric acid compared n with C. glabrata. ϭ 50) Boric acid or boracic [B(OH)3]isa weak acid, and its mode of antifungal ac- tion is not clear. Shiohara and Tasker (15) proposed that its acidic properties lead to dsratio Odds disruption of the fungal cell wall. The pH CI) (95% of boric acid–Sabourauds’s broth (5.0– 5.09) required for in vitro inhibition of C. albicans growth has been reported to be similar to the vaginal pH in untreated Candida vaginitis, indicating that the ef- P

fect of boric acid is not speciﬁc to its acidic value properties (17). Alkaline compound (i.e., Na2B4O3 [borax]) also has antiseptic

properties (32). The doses of the boric kerke acid used to treat VVC have ranged from Nagel-

600 mg daily for 7 days (16) to twice daily R for 14 days (18). We used 600 mg boric 2

DIABETES CARE, VOLUME 30, NUMBER 2, FEBRUARY 2007 315 Diabetes and vulvovaginal candidiasis acid daily for 14 days because Ͼ90% my- in Candida glabrata: coordinate upregu- Acknowledgments— The authors are thank- cological cure has been reported with this lation of multidrug transporters and evi- ful to the financial support by ICMR, New dence for a Pdr1-like transcription factor. schedule in nondiabetic patients with Delhi, for this study. symptomatic C. albicans (17) or C. gla- Antimicrob Agents Chemother 48:3773– 3781, 2004 brata VVC (14). 14. Sobel JD, Chaim W: Treatment of Toru- Boric acid therapy (600 mg/day for lopsis glabrata vaginitis: retrospective re- 14 days) has been found to be safe References view of boric acid therapy. Clin Infect Dis (14,17), except for local burning sensa- 1. Sonck CE, Somersalo O: The yeast flora of 24:649–652, 1997 tion and vestibular erythema in occa- anogenital region in diabetic girls. Arch 15. Shinohara YT, Tasker SA: Successful use Dermatol 83:214–220, 1963 sional patients (14,17,18). Two of our of boric acid to control azole-refractory 2. Reed BD: Risk factors for Candida vulvo Candida vaginitis in a woman with AIDS. patients also reported this side effect. vaginitis. Obstet Gynecol Surv 47:551–560, Improvement in clinical features of VVC J Acquir Immune Defic Syndr Hum Retrovi- 1992 rol 16:219–220, 1997 was comparable in two treatment arms. 3. Peer AK, Hoosen AA, Seedat MA, van den 16. Swate TE, Weed JC: Boric acid treatment Ende J, Omar MA: Vaginal yeast infec- However, we cannot exclude an ele- of vulvovaginal candidiasis. Obstet Gy- tions in diabetic women. S Afr Med J 83: ment of bias in the analysis of this necol 43:893–895, 1974 727–729, 1993 secondary outcome, as the same inves- 17. Van Slyke KK, Michel VP, Rein MF: Treat- 4. Goswami R, Dadhwal V, Tejaswi S, Datta tigator carried out the clinical examina- ment of vulvovaginal candidiasis with bo- K, Paul A, Haricharan RN, Banerjee U, ric acid powder. Am J Obstet Gynecol 141: tion following treatment. Kochupillai NP: Species-specific preva- 145–148, 1981 In our previous studies, poor glyce- lence of vaginal candidiasis among pa- 18. Jovanovic R, Congema E, Nguyen HT: mic control was linked with Candida tients with diabetes mellitus and its Antifungal agents vs. boric acid for treat- growth (4) but was unrelated to the type relation to their glycaemic status. J Infect ing chronic mycotic vulvovaginitis. J Re- 41:162–166, 2000 of species isolated or response to the flu- prod Med 36:593–597, 1991 conazole therapy (8). In the current 5. Bohannon NJ: Treatment of vulvovaginal candidiasis in patients with diabetes. Di- 19. Expert Committee on the Diagnosis and study, there was a trend of higher mean Classification of Diabetes Mellitus: Report A1C levels in those showing persistence abetes Care 21:451–456, 1998 6. de Leon EM, Jacober SJ, Sobel JD, Fox- of the Expert Committee on the Diagnosis of Candida growth following either of the man B: Prevalence and risk factors for vag- and Classification of Diabetes Mellitus. two therapies. The cause of increased C. inal Candida colonization in women with Diabetes Care 20:1183–1197, 1997 glabrata isolation in diabetic women is not type 1 and type 2 diabetes. BMC Infect Dis 20. Eckert LO, Hawes SE, Stevens CE, clear but may involve frequent use of an- 2:1, 2002 Koutsky LA, Eschenbach DA, Holmes KK: tifungal drugs leading to its reduced sus- 7. Jurevic RJ, Bai M, Chadwick RB, White Vulvovaginal candidiasis: clinical mani- TC, Dale BA: Single-nucleotide polymor- festations, risk factors, management algo- ceptibility to azoles (33) and consequent rithm. Obstet Gynecol 92:757–765, 1998 polarization/homing in diabetic women. phisms (SNPs) in human beta-defensin 1: high-throughput SNP assays and associa- 21. Fotedar R, Banerjee U: Changing pattern of Feng et al. (34) reported lesser suscepti- tion with Candida carriage in type I dia- Candida species in a bone marrow trans- bility of C. glabrata in comparison with C. betics and nondiabetic controls. J Clin plant patient. J Infect 32:243–245, 1996 ␤ albicans to -defensins, natural cationic Microbiol 41:90–96, 2003 22. Guidelines on Standard Operating Proce- antimicrobial/antifungal peptides ex- 8. Goswami D, Goswami R, Banerjee U, dures for Laboratory Diagnosis of HIV-Op- pressed in human epithelia. In diabetic Dadhwal V, Miglani S, Lattif AA, Koch- portunistic Infection. Kumari S, Ed. New milieu, ␤-defensins expression is reduced upillai N: Pattern of Candida species iso- Delhi, World Health Org. Regional Office lated from patients with diabetes mellitus for South East Asia (35). Reduced expression of defensins in 23. Hosmer DW Jr, Lemeshow S: Applied Lo- association with resistance of C. glabrata and vulvovaginal candidiasis and their response to single dose oral fluconazole gistic Regression. New York, John Wiley & to fungicidal activity of drugs like flucon- therapy. J Infect 52:111–117, 2006 Sons, 1989 azole may also explain the high preva- 9. Corsello S, Spinillo A, Osnengo G, Penna 24. Geiger AM, Foxman B, Sobel JD: Chronic lence of C. glabrata VVC in diabetic C, Guaschino S, Beltrame A, Blasi N, Festa vulvovaginal candidiasis: characteristics women. A: An epidemiological survey of vulvo- of women with Candida albicans, C gla- One of the relevant findings in this vaginal candidiasis in Italy. Eur J Obstet brata and no candida. Genitourin Med 71: study is the importance of species identi- Gynecol Reprod Biol 110:66–72, 2003 304–307, 1995 fication of Candida isolates for proper 10. Redondo-Lopez V, Lynch M, Schmitt C, 25. Horowitz BJ, Edelstein SW, Lippman L: Candida tropicalis vulvovaginitis. Obstet management of VVC in diabetic women. Cook R, Sobel JD: Torulopsis glabrata vaginitis: clinical aspects and susceptibil- Gynecol 66:229–232, 1985 It is emphasized that conventional inex- ity to antifungal agents. Obstet Gynecol 76: 26. Odds FC: Resistance of yeasts to azole- pensive testing procedures, though time 651–655, 1990 derivative antifungals. J Antimicrob Che- consuming, are good enough to identify 11. vanden Bossche H, Marichal P, Odds FC, mother 31:463–471, 1993 Candida up to species level and would Le Jeune L, Coene MC: Characterization 27. Dennerstein GJ: Depo-Provera in the help in the management of diabetic of an azole-resistant Candida glabrata iso- treatment of recurrent vulvovaginal can- women with VVC. As an alternative, boric late. Antimicrob Agents Chemother 36: didiasis. J Reprod Med 31:801–803, 1986 acid therapy could be considered as the 2602–2610, 1992 28. Sanglard D, Ischer F, Calabrese D, Ma- jcherczyk PA, Bille J: The ATP binding frontline therapy for treating VVC in dia- 12. Fidel PL Jr, Vazquez JA, Sobel JD: Can- dida glabrata: review of epidemiology, cassette transporter gene CgCDR1 from- betic women because it is effective against pathogenesis, and clinical disease with Candida glabrata is involved in the resis- both C. albicans and C. glabrata compared comparison to C. albicans. Clin Microbiol tance of clinical isolates to azole with fluconazole, which is effective Rev 12:80–96, 1999 antifungal agents. Antimicrob Agents Che- against C. albicans only. 13. Vermitsky JP, Edlind TD: Azole resistance mother 43:2753–2765, 1999

316 DIABETES CARE, VOLUME 30, NUMBER 2, FEBRUARY 2007 Ray and Associates

29. Gonzalez Ortiz M, Martinez Abundis E: cies isolated from female prostitutes with 7:495–504, 2006 Itraconazole in the treatment of Candida vulvovaginitis to antifungal agents and 34. Feng Z, Jiang B, Chandra J, Ghannoum M, vulvovaginitis in patients with type II di- boric acid. Eur J Clin Microbiol Infect Dis Nelson S, Weinberg A: Human beta-de- abetes mellitus (non-insulin dependent) 18:59–61, 1999 fensins: differential activity against can- Ginecol Obstet Mex 63:15–18, 1995 32. Carey H: Double blind clinical trial of Bo- didal species and regulation by Candida 30. Balbi C, D’Ajello M, Balbi GC: Treatment rax and Candida in the treatment of vag- albicans. J Dent Res 84:445–450, 2005 with ketoconazole in diabetic patients inal discharge. Comm Br Homoeopath Res 35. Froy O, Hananel A, Chapnik N, Madar Z: with vaginal candidiasis. Drugs Exp Clin Grp 15:12–14, 1986 Differential effect of insulin treatment on Res 12:413–414, 1986 33. Ruhnke M: Epidemiology of Candida al- decreased levels of beta-defensins and 31. Otero L, Fleites A, Mendez FJ, Palacio V, bicans infections and role of non-Candi- Toll-like receptors in diabetic rats. Mol Vazquez F: Susceptibility of Candida spe- da-albicans yeasts. Curr Drug Targets Immunol 44:796–802, 2007

DIABETES CARE, VOLUME 30, NUMBER 2, FEBRUARY 2007 317