S124

Is Vulvovaginal Candidiasis an AIDS-Related Illness?

Mary H. White From the Infectious Disease Service, Memorial Sloan-Kettering
Cancer Center, New York, New York

Early in the AIDS epidemic, retrospective studies reported that vaginal candidiasis occurred more
frequently in women who were infected with human immunodeficiency virus (HIV) than in those
who were not infected. Some investigators suggested that new onset or recurrent vaginal candidiasis
might identify HIV-infected individuals and predict the course of AIDS in women already known
to be infected. In this article, studies of vaginal candidiasis in HIV-infected women are examined, and

Downloaded from https://academic.oup.com/cid/article/22/Supplement_2/S124/478526 by guest on 10 January 2022

several observations are made. First, early studies were small and likely reflected biased populations.
Second, adherence to previously accepted diagnostic criteria for vaginal candidiasis was not consis-
tent in these studies. Finally, conclusions about the increased risk of recurrent or chronic candidal
vaginitis in HIV-infected women have been promulgated in the medical literature and may have
influenced clinical practice even though such statements are not supported epidemiologically. Pro-
spective trials with uninfected community controls should determine the true impact of HIV infection
on vulvovaginal candidiasis.

Much of the interest in gender-specific illness associated
with AIDS has resulted from the concern that HIV-infected
women are underrecognized [1]. The consequences of delayed
recognition are lack of benefit from primary antiretroviral and
prophylactic treatments, which may translate to decreased sur-
vival rates [2]. Identification of female-specific stigmata may
prove useful in the early recognition of HIV disease in women.
However, it is important to avoid overinterpretation of small
or preliminary reports, as unnecessary concern is raised when
drawing conclusions about recommendations for testing for
HIV antibodies in women whose infection status is unknown
[3].
An example is the significance ofvulvovaginal candidiasis in
HIV-infected women. An increase in the frequency of candidal
vaginitis has been linked with both progression of known HIV
disease and a greater likelihood of having HIV infection itself
[4]. In this article, studies of vaginal candidiasis in HIV-
infected women are reviewed, and questions about conclusions
drawn from initial studies are raised.
Esophageal Candidiasis: An Instructive History
Both esophageal candidiasis and candidal vaginitis are mu-
cosal infections treated with topical and systemic antifungal
agents on an intermittent or continuous basis. Early in the AIDS
epidemic, both illnesses were reported to occur with greater
frequency in HIV-infected women [4, 5]. The following brief
review examines the progress in our understanding of candidal
esophagitis; this review may be useful when interpreting studies
of vaginal candidiasis in HIV-infected individuals.
In an early study of AIDS in 24 women in Rhode Island by
Carpenter et al. [5], candidal esophagitis was the most frequent
AIDS-defining infection (38% of the women). In addition,
Pneumocystis carinii pneumonia (PCP) was noted in only 13%
of the women. These findings differed considerably from the
reported incidence of opportunistic infections in patients with
AIDS elsewhere [6], thus prompting Carpenter et al. to suggest
that the course of HIV infection differed in women and men.
More recent studies [7-9] demonstrated no gender differ-
ences in the incidence of candidal esophagitis and other clinical
manifestations of HIV infection. It is likely that early small
studies selected for female patients with advanced HIV infec-
tion or atypical presentations because of lack of recognition of
women's risk factors for HIV infection. For example, a missed
diagnosis of PCP in a woman would lead to death, thus pre-
venting her from being treated by referral physicians. In con-
trast, a chronic nonfatal illness, such as candidal esophagitis,
would be more readily diagnosed and prompt testing for HIV
antibodies.
An improved understanding of the incidence of esophageal
candidiasis resulted from the implementation of large, con-
trolled, community-based trials. This example of improved un-
derstanding should guide future planning for studies ofcandidal
vaginitis in HIV-infected women.

Financial support: This work was supported in part by the National Institute Vulvovaginal Candidiasis Before the AIDS Epidemic
of Allergy and Infectious Diseases (AI-25917) and the Charles Lawrence Keith
and Clara Miller Foundation.
Long before the AIDS epidemic, vulvovaginal candidiasis
Reprints or correspondence: Dr. Mary H. White, Infectious Disease Service, was well studied, and its clinical definitions were established
Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, New [10]. Colonization was described in ,....,20% of women in point
York 10021.
prevalence studies before the AIDS epidemic began [11].
Clinical Infectious Diseases 1996; 22(Suppl 2):SI24-7
© 1996 by The University of Chicago. All rights reserved.
Acute vulvovaginal candidiasis is a clinical syndrome that
1058--4838/96/2205-0007$02.00 includes perineal pruritus, white curdlike discharge, dyspareu-
cm 1996; 22 (Suppl 2) Vaginal Candidiasis in HIV-Infected Women
Table 1. Potential host factors associated with the pathogene-
sis of recurrent vulvovaginal candidiasis.
Hormonal factors
Pregnancy
Estrogens
Corticosteroids
Oral contraceptives
Local factors
Tight fitting and synthetic clothing
Local allergy (commercial douches, perfumes)
Immunologic factors
HIV infection!AIDS
Idiopathic or other immune deficiency
Other factors
Antibiotics
Diabetes mellitus
nia, vulvar erythema, and dysuria; potassium hydroxide stain-
ing of specimens from patients with this infection reveals
yeasts, and culture yields Candida species. In one study [12],
acute vaginal candidiasis was observed in 20% of20,000 symp-
tomatic women examined in an office setting over as-year
period.
The importance of following both clinical and microbiologi-
cal diagnostic criteria was pointed out by Adler et al. [13] and
Adler and Belsey [14] who found that only 4% of episodes of
vaginal candidiasis were diagnosed correctly on clinical
grounds alone. In another study, Berg et al. [15] examined 134
women with genitourinary symptoms and 70 asymptomatic
women. The frequency of isolation of Candida species from
vaginal cultures did not differ between the two groups.
Recurrent vulvovaginal candidiasis also predated the AIDS
epidemic. Recurrent vulvovaginal candidiasis is defined as at
least four microbiologically proven symptomatic episodes of
acute vaginal candidiasis within a 12-month period in the ab-
sence of other causes of vaginal symptomatology, including
trichomoniasis, bacterial vaginosis, and allergic and!or chemi-
cal reactions to topical treatment [16].
The etiology of recurrent vulvovaginal candidiasis has been
explored extensively. Table 1 summarizes several possible host
factors that may predispose to recurrent vulvovaginal candidia-
sis [12, 16]. However, most women with recurrent vaginitis
have no definable underlying condition [17].
Endogenous reinfection, particularly from a gastrointestinal
reservoir, has been implicated in the pathogenesis of recurrent
vulvovaginal candidiasis, but attempts at reducing the number
of vaginitis events by clearing yeasts from the gastrointestinal
tract have been unsuccessful [18]. Sexual partners of women
with recurrent vulvovaginal candidiasis have also been impli-
cated since 20% to 30% of them have penile carriage of Can-
dida [19]. Treatment of male partners did not diminish the
frequency of vaginitis in a prospective trial [20], but redesign
of such trials with a greater emphasis on the male partner's
sexual activity, history of circumcision, medication compli-
8125
ance, and results of serial fungal cultures may be necessary to
clarify the impact of sexual activity on recurrent vulvovaginal
candidiasis.
Vazquez et al. [21] proposed that local recolonization may
playa role in the pathogenesis of recurrent vulvovaginal candi-
diasis. Using the CHEF (contour-clamped homogeneous elec-
tric field electrophoresis) technique over a 3-year observation
period, they demonstrated that eight of 10 women with recur-
rent vulvovaginal candidiasis were infected with the same
karyotypically identical strain that had infected them at the
Downloaded from https://academic.oup.com/cid/article/22/Supplement_2/S124/478526 by guest on 10 January 2022
onset of study.
Although the pathogenesis of acute and recurrent vulvovagi-
nal candidiasis has not been elucidated, the prevalence and
incidence of vaginal colonization with Candida and acute can-
didal vaginitis in women were well established before the AIDS
era. Observations regarding HIV-infected women must be in-
terpreted with this background in mind.
Candidal Vaginitis as an AIDS-Associated Illness
Table 2 summarizes data from initial studies of vaginal can-
didiasis in women with AIDS. In the study by Rhoads et al.
[22], seven women who presented with a complaint of chronic
vaginal candidiasis during the prior year were described. On
the basis of their observations, Rhoads et al. concluded the
following: recurrent vulvovaginal candidiasis is an indicator of
HIV infection, it suggests severe underlying immunodefi-
ciency, and it is associated with a poor prognosis (because
HIV-infected women with recurrent vulvovaginal candidiasis
had AIDS and died sooner than did HIV-infected women with-
out recurrent vulvovaginal candidiasis).
These conclusions may have been premature for several rea-
sons. First, the investigators correctly defined candidal vagini-
tis, but the diagnoses for their patients appear to have been
based on patient history, since a I-year history of vaginitis
was their complaint at the time of presentation. As mentioned
previously, clinical symptoms alone are unreliable for making
a diagnosis of vaginal candidiasis [13 -15]. Second, documen-
tation by culture and elimination of other pathogens as the
cause of symptoms were not reported. Third, no microbiologi-
cal comparisons were made between these seven patients and
Table 2. Summary of data from early studies of HIV-infected
women with chronic or recurrent vulvovaginal candidiasis.
No. (%) with chronic or
recurrent vulvovaginal
[Reference] year Study population candidiasis
[22] 1987 29 HIV-positive women 7 (24)
[4] 1990 66 HIV-positive women 33 (50)
[23] 1991 117 symptomatic HIV- 43 (37)
positive women
[23] 1991 200 HIV-positive women 89 (45)
S126
22 other HIV-infected women without vaginal symptoms or
uninfected community controls. Fourth, Rhoads et al. noted
that all of the women with vaginitis also had oral thrush. In
previous reports, oral thrush independently predicted progres-
sion to AIDS [24]; this fact may explain the poor outcome for
the female patients with vaginitis.
Imam et al. [4], who observed 66 women with HIV infection,
defined HIV-associated vaginal candidiasis as a doubling in
the number of vaginitis episodes over the previous year. By
this definition, 33 patients (50%) had new onset or recurrent
vaginal candidiasis. This study also reported that a hierarchical
pattern of mucosal candidal infections exists: i.e., the incidence
of vaginal, oral, and esophageal candidiasis was inversely pro-
portional to the CD4 lymphocyte count. Vaginal infections
occurred in women with near-normal counts, and esophagitis
occurred in women with <50 CD4+ lymphocytes/mm3 .
In another report [23], candidal vaginitis was reported as the
most frequent opportunistic infection, and recurrent candidal
vaginitis was the most frequent initial clinical manifestation in
women with HIV infection. The investigators concluded that
women should consider testing for HIV antibodies if they have
experienced more than four annual episodes of vaginal candidi-
asis or a doubling in the number of episodes from the previous
year [4, 23].
New onset or recurrent vulvovaginal candidiasis or an in-
creased frequency of recurrent vulvovaginal candidiasis has
not been epidemiologically proven to be associated with AIDS,
nor do these findings satisfy the case definition for AIDS [25].
In addition, none of the three studies had a control population
of HIV-negative women who engaged in similar risk behaviors
in the same community. With regard to the hierarchical pattern
of candidal infection, the data reflected only a cohort of women
who reported vaginal, oral, and esophageal infections as their
initial manifestations of candidiasis. The results do not reflect
the longitudinal incidence among a cohort of individuals.
Nonetheless, the conclusions drawn from these studies may
have influenced clinical practice, because decisions about test-
ing for HIV antibodies and therapy for women known to be
HIV-infected may be based on the occurrence of new onset
vaginal candidiasis or on an increased frequency of this infec-
tion (M. H. White, unpublished data and [3]).
Candida) Vaginitis as a Non-AIDS-Associated Illness
Investigators have attempted to determine the prevalence of
vaginal colonization in HIV-infected women as a surrogate for
candidal vaginitis. At Memorial Sloan-Kettering Cancer Center
(New York), 50 HIV-infected women were followed up by
routine physical examination over a 30-month period; Candida
species were isolated during 26% of visits (M. H. White, un-
published data), a prevalence not dissimilar to that among his-
torical controls. Another study [26] compared vaginal coloniza-
tion with CD4+ lymphocyte counts in pregnant HIV-infected
women at 34 weeks' gestation and 2 months after delivery.
White em 1996; 22 (Suppl 2)
Table 3. Principles in the study of genital candidiasis in HIV-
infected women.
Diagnosis
Distinguish colonization from infection
Exclude other causes of vaginal symptomatology
Satisfy definition for acute or recurrent vulvovaginal candidiasis
Epidemiology
Compare rate of colonization and incidence of infection with
those among community controls
Determine incidence of symptomatic vaginal candidiasis in
colonized individuals
Downloaded from https://academic.oup.com/cid/article/22/Supplement_2/S124/478526 by guest on 10 January 2022
Determine predictors or risk factors for the development of
symptomatic candidiasis
Control for putative risk factors of vaginal candidiasis, e.g.,
antibacterial antibiotics
The prevalence of colonization was inversely related to the
absolute CD4+ lymphocyte count. In a similar study with non-
pregnant HIV-infected women [27], the incidence of recurrent
vulvovaginal candidiasis increased with declining CD4+ lym-
phocyte counts.
In none of these three studies was a concurrent community
control group used for comparison. In addition, in the latter
two studies, a relationship between colonization or vaginitis
and therapy or prophylaxis with antibacterial antibiotics was
not examined; this is an important factor for patients with
<200 CD4+ lymphocytes/mm3 . The increased frequency of
colonization and recurrent vulvovaginal candidiasis in such
patients may be explained by these and other unidentified fac-
tors, rather than by the specific immune defect of HIV infection.
Two controlled studies [28, 29] were recently reported.
In the first study [28], which took place in New York City,
several hundred HIV-infected and uninfected women from
the same community were followed; no difference in the
prevalence of vaginal colonization with Candida was found.
The other study [29] compared rates of oral, vaginal, and
rectal colonization with Candida in HIV-infected women
and community controls in four cities in the United States.
Overall colonization rates were statistically increased among
HIV-infected women, but no relationship was established
between vaginal colonization and absolute CD4 + lympho-
cyte counts in these individuals.
It is not known if the incidence of vaginitis in HIV-in-
fected women and uninfected women differs, as this rate has
not been studied in controlled trials. It is important to note
that even if future trials do not demonstrate an increased
incidence of acute and recurrent vulvovaginitis in HIV-in-
fected women as a whole, the incidence of vaginitis in sub-
groups-women with <200 CD4+ lymphocytes/mm3 (who
receive prophylaxis with antibiotics), women receiving che-
motherapy or corticosteroid therapy, and women who are
pregnant-might be higher. Table 3 summarizes some of
the principles that current and future studies should consider
to clarify some of these issues.
CID 1996;22 (Suppl 2) Vaginal Candidiasis in HIY-Infected Women
Conclusions
Results from early studies of HIV-infected women suggested
that vaginal candidiasis was more frequent and more likely to
recur in HIV-infected women than in uninfected women. These
results prompted the recommendation for testing for HIV anti-
bodies in women with new onset vaginal candidiasis or an
increased frequency of this infection. Certainly, any woman
with a history of intravenous drug use and/or unprotected sex-
ual activity should consider testing for HIV antibodies. In addi-
tion, unusual or refractory manifestations ofcommon infections
should lead to an evaluation for immunodeficiency.
However, candidal vaginitis has not been proven to be more
frequently associated with HIV infection. Current studies are
under way, and the results from these investigations should be
interpreted in light of the principles outlined earlier. Until then,
recommendations for HIV antibody testing in women not
known to be infected with HIV and for therapeutic interven-
tions in women known to be infected with HIV should not be
based on the diagnosis of candidal vaginitis alone.
References
1. Schoenbaum EE, Webber MP. The underrecognition of HIV infection in
women in an inner-city emergency room. Am J Public Health
1993;83:363-8.
2. Lemp GF, Hirozawa AM, Cohen JB, Derish PA, McKinney KC, Hernan-
dez SR. Survival for women and men with AIDS. J Infect Dis
1992; 166:74-9.
3. DeHart DJ. HIV testing in women with vaginal candidiasis. Am J Med
1991; 90:536.
4. Imam N, Carpenter CCJ, Mayer KH, Fisher A, Stein M, Danforth SB.
Hierarchical pattern of mucosal Candida infections in HIV-seropositive
women. Am J Med 1990;89:142-6.
5. Carpenter CCJ, Mayer KH, Fisher A, Desai MB, Durand L. Natural history
of acquired immunodeficiency syndrome in women in Rhode Island.
Am J Med 1989;86:771-5.
6. New York City Department of Health AIDS Surveillance. The AIDS
epidemic in New York City, 1981-1984. Am J Epidemiol
1986; 123:1013-25.
7. Morlat P, Parneix P, Douard D, et al. Women and HIV infection: a cohort
study of 483 HIV-infected women in Bordeaux, France, 1985 -1991.
AIDS 1992;6:1187-93.
8. Phillips AN, Antunes F, Stergious G, et al. A sex comparison of rates
of new AIDS-defining disease and death in 2554 AIDS cases. AIDS
1994;8:831-5.
9. Vlahov D, Munoz A, Solomon L, et al. Comparison of clinical manifesta-
tions of HIV infection between male and female injecting drug users.
AIDS 1994;8:819-23.
10. MonifGRG. Classification and pathogenesis of vulvovaginal candidiasis.
Am J Obstet Gynecol 1985; 152:935-9.
SI27
II. Odds Fe. Ecology of Candida and epidemiology of candidosis. In: Odds
FC, ed. Candida and candidosis: a review and bibliography. 2nd ed.
London: Balliere Tindall, 1988:68-92.
12. Fleury FJ. Adult vaginitis. Clin Obstet Gynecol 1981;24:407-38.
13. Adler MW, Belsey EM, Rogers JS. Sexually transmitted diseases in a
defined population of women. Br Med J 1981;283:29-32.
14. Adler M, Belsey E. The G.P. and the specialist: gynaecology. Br Med J
1983;286:890.
15. Berg AO, Heidrich FE, Fihn SD, et al. Establishing the cause of genitouri-
nary symptoms in women in a family practice. JAMA 1984; 251:
620-5.
16. Sobel JD. Pathogenesis and treatment of recurrent vulvovaginal candidia-
Downloaded from https://academic.oup.com/cid/article/22/Supplement_2/S124/478526 by guest on 10 January 2022
sis. Clin Infect Dis 1992; 14(suppl 1):SI48-53.
17. Forssman L, Milsom J. Treatment of recurrent vaginal candidiasis. Am J
Obstet Gynecol 1985; 152:959-61.
18. Milne JD, Warnock OW. Effect of simultaneous oral and vaginal treatment
on the rate of cure and relapse in vaginal candidiasis. Br J Vener Dis
1979;55:362-6.
19. Rodin P, Kolator B. Carriage of yeasts on the penis. Br Med J
1976; 1:1123-4.
20. Sobel JD. Management of recurrent vulvovaginal candidiasis with intermit-
tent ketoconazole prophylaxis. Obstet GynecoI1985;65:435-40.
21. Vazquez JA, Sobel JD, Demitriou R, Vaishampayan J, Lynch M, Zervos
MJ. Karyotyping of Candida albieans isolates obtained longitudinally
in women with recurrent vulvovaginal candidiasis. J Infect Dis
1994; 170:1566-9.
22. Rhoads JL, Wright C, Redfield RR, Burke OS. Chronic vaginal candidiasis
in women with human immunodeficiency virus infection. JAMA
1987;257:3105-7.
23. Carpenter CCJ, Mayer KH, Stein MO, Leibman BD, Fisher A, Fiore TC.
Human immunodeficiency virus infection in North American women:
experience with 200 cases and a review of the literature. Medicine
(Baltimore) 1991; 70:307 -25.
24. Klein RS, Harris CS, Small CB, et al. Oral candidiasis in high-risk patients
as the initial manifestation of the acquired immunodeficiency syndrome.
N Engl J Med 1984;311:354-8.
25. Centers for Disease Control and Prevention. 1993 revised classification
system for HIV infection and expanded surveillance case definition for
AIDS among adolescents and adults. MMWR Morbid Mortal Wkly Rep
1992;41(RR-17):1-19.
26. Burns 0, Tuomala R, Regan J, et al. Positive vaginal cultures for
C. albieans correlate with C04+ level among pregnant and postpartum
HIY-l-infected women [abstract no 1676]. In: Program and abstracts
of the 33rd Interscience Conference on Antimicrobial Agents and Che-
motherapy (New Orleans). Washington, DC: American Society for Mi-
crobiology, 1993:425.
27. Clark R, Brandon W, Blakley S, Rice J. Clinical manifestations in HIV +
women [abstract no 1329]. In: Program and abstracts of the 33rd Inter-
science Conference on Antimicrobial Agents and Chemotherapy (New
Orleans). Washington, DC: American Society for Microbiology,
1993:363.
28. Ellerbrock T, Wright T, Rice R, Chiasson MA. Genital tract infections in
HIV-infected women [abstract no FCI-180]. In: Program and abstracts
of the 1st HIY Infection in Women Conference (Washington, DC).
Richmond, VA: Philadelphia Sciences Group, 1995.
29. Schuman P, Sobel JD, Mayer K, et al. Candida colonization in women at
risk for HIV infection [abstract no FC 1-176]. In: Program and abstracts
of the 1st HIV Infection in Women Conference (Washington, DC).
Richmond, VA: Philadelphia Sciences Group, 1995.