Kaposi's sarcoma (KS) is best known in the United States in association
with acquired immunodeficiency syndrome (AIDS). Worldwide, however,
the AIDS-related form is only one of four clinical presentations
of the disease. Much remains unknown about the pathogenesis of
AIDS-related KS. Some aspects of the epidemiology of the four
types, as well as the changing picture of AIDS-related KS, may
provide important leads to understanding the etiology(ies) of
The four types of KS are the classic form, the African form, the
form associated with iatrogenic immunosuppression, and the AIDS-related
form. The classic form of KS was first described as "idiopathic,
multiple pigmented sarcomas of the skin" in 1872 by Moritz
Kaposi, a Viennese dermatologist". Kaposi's description
was based on the examination of three elderly men who had lesions
on their extremities. Since the publication of Kaposi's original
paper, this form of KS has been observed worldwide, although it
seems to be more prevalent in people of Mediterranean origin.
It is not rapidly progressive, and the male-to-female ratio is
10 to 15:1. A typical patient might present with KS in his 60s
and live another 10 or 15 years before dying of some unrelated
The African endemic form of KS was first described in 1914, but
its prevalence in Africa was only appreciated approximately 20
years later, when a review of 500 tumors in Nigerian patients
showed 2% to be KS [3,4]. The African form occurs predominantly
in young, black adults 25 to 40 years of age. The male-to-female
ratio for this form is 17:1. The presentation is variable. Benign,
nodular lesions limited to the extremities form one end of the
spectrum; florid, disseminated lesions are at the other end. Survival
ranges from 3 to 10 years following presentation. There is also
a lymphadenopathic subvariant of the African form that affects
children at a mean age of 3 years. The male-to-female ratio for
this pediatric subvariant is 3:1 .
In the 1970s, a third form of KS was observed in immunocompromised
patients. Such patients who are at risk for this form of KS include
recipients of renal transplant, patients on long-term corticosteroids
for various disorders, and patients who are immunosuppressed as
a result of some other therapeutic regimen or malignancy. As with
the African form, with the immunocompromised form, lesions are
often localized, although rapid dissemination is seen in some
cases. The male-to-female ratio for this form is 2.3:1. One unique
aspect of this form of KS is that some lesions resolve when iatrogenic
immunosuppression is stopped .
In 1981, a major epidemic of the fourth type of KS was discovered.
The AIDS-related form occurs in conjunction with human immunodeficiency
virus (HIV) infection and is found predominantly among white,
homosexual/bisexual men. In contrast to the other three forms,
in which lesions tend to appear on the legs, AIDS-related KS may
appear on the extremities but also commonly appears on the chest
and face (Figure 1) as well as in the gastrointestinal tract and
lungs. Disease is also more virulent with this form than with
the other three forms, and most patients die of KS or other AIDS-defining
illnesses within 1 to 3 years of presentation. The male-to-female
ratio for the AIDS-related form is 100:1 .
There may be some confusion about presentation when forms of KS
intersect. For example, as the AIDS epidemic spreads in Africa,
AIDS-related KS is overtaking the endemic African form in frequency.
Researchers from Zambia report that the prevalence of KS is increasing
in tandem with that of HIV infection. The male-to-female ratio
for all cases of KS dropped from 10:1 in 1985 to 2.3:1 in 1993.
The HIV seroprevalence in Zambian cases of KS is 97% . Another
example is the juxtaposition of HIV-positive and HIV-negative
patients with KS among homosexual men in the United States. Compared
with HIV-positive patients, HIV-negative patients are older (median
age, 49 years vs 37 years), less immunosuppressed (median CD4
counts, 768 vs 226 cells/mm³), and have more indolent clinical
courses (less than10 skin lesions for all HIV-negative patients)
In 1981, the Centers for Disease Control and Prevention (CDC)
in Atlanta, Georgia, initiated surveillance of AIDS in the United
States. Between June 1981 and December 31, 1992, a total of 253,322
cases of patients with AIDS were reported. Presumptive or definitive
KS was reported in 33,170 patients (13%). There are three caveats
regarding the interpretation of these figures. First, the definition
of AIDS and AIDS-related KS changed several times during the surveillance
process. Second, approximately 20% of cases of KS are diagnosed
without benefit of biopsy; this is particularly common in patients
with other manifestations of AIDS, who are often diagnosed with
KS on clinical grounds alone. Third, there are cases of KS that
go unreported when patients with AIDS develop KS in the later
stages of disease, after AIDS has already been reported .
Despite these limitations, several observations can be made from
the surveillance data. First, homosexual/bisexual men are much
more likely to develop KS than are patients from other groups
Table 1). More than 30,000 of the 33,170 AIDS-related cases of
KS have occurred among homosexual/bisexual men, a rate of KS of
19.2% in the homosexual/bisexual patient population with AIDS.
In other groups, the rate ranges from 0.5% to 5.6%. This difference
is highly significant.
Second, the incidence of AIDS-related KS increased each year from
1981 through 1990. However, unlike total cases of AIDS, which
have continued to increase, the incidence of AIDS-related KS declined
in 1991 and 1992. In fact, the proportion of cases of AIDS with
KS has declined each year since surveillance began (
Third, among blood transfusion-related cases of AIDS, 42 donor-recipient
pairs developed AIDS. There were no cases in which the recipient
developed KS after receiving transfusion from donors who later
developed KS. In one case, a recipient developed KS after receiving
HIV-infected blood from a donor who did not develop KS. If AIDS-related
KS was caused by HIV alone or a special strain(s) of HIV, some
concordance of KS between donors and recipients would be expected
Fourth, among homosexual/bisexual men with AIDS, KS is more common
in white men than in black men (22% vs 10%) [7,8].
Fifth, initial data showed that women who were sexual partners
of homosexual men were more likely to develop KS than women sexually
linked to injection drug users (3% vs 0.9%) . The gap between
these groups is narrowing, however, and is no longer statistically
Table 3) . Women who reported intravenous drug
abuse or a history of blood transfusion prior to the diagnosis
of AIDS were excluded from analysis.
Finally, the rate of KS among patients with AIDS varies with location.
The proportion of white, homosexual patients with AIDS and KS
is more than 20% in California, Connecticut, the District of Columbia,
Massachusetts, New Mexico, New York, and South Carolina but less
than 10% in Alabama, Arkansas, Iowa, Nevada, and Utah .
The unique epidemiology of KS among patients with AIDS suggests
that its cause is multifactorial. Although HIV plays a major role
in its pathogenesis, HIV alone is not its cause. Data suggest
that a necessary factor(s), and/or enhancing factor(s), associated
with a homosexual life-style predisposes HIV-infected persons
to KS. A second sexually transmitted organism, a fecal-orally
transmitted organism, genetics, and nitrite inhalant use have
been proposed as cofactors. The current epidemic of AIDS-related
KS provides a unique opportunity to decipher the pathogenesis
of a malignancy that is multifactorial in etiology.
1. Kaposi M: Idiopathisches multiples Pigmentsarkom der Haut.
Arch Dermatol Syph 4:265-273, 1872.
2. Friedman-Kien AE: Color Atlas of AIDS. Philadelphia, WB Saunders
3. Hallenberger O: Multiple Angiosarkome der Haut bei einem Kamerunneger.
Arch Schiffs Trop Hyg 18:647-651, 1914.
4. Smith EC, Elmes BGT: Malignant disease in natives of Nigeria:
An analysis of 500 tumours. Ann Trop Med Parasitol 28:461-513,
5. Ebrahim SH, Sunkutu MR, Mwansa N: Epidemiological and clinical
features of Kaposi's sarcoma in adults in Zambia (abstract WS-B15-1).
Proceedings of the IX International Conference on AIDS. Berlin,
Germany, June 6-11, 1993.
6. Saltzman BR, Friedman-Kien AE, Huang YQ, et al: Epidemic Kaposi's
sarcoma in HIV uninfected (HIV-neg) homosexual (HS) men: Clinical
course and analysis for known human retroviral genomic sequences
by PCR (abstract ThB 1543). Proceedings of the XIII International
Conference on AIDS. Amsterdam, Netherlands, July 19-24, 1992.
7. Haverkos HW, Drotman DP: Epidemiology of AIDS-related Kaposi's
sarcoma (KS): An update (abstract PO-B12-1576). Proceedings of
the IX International Conference on AIDS. Berlin, Germany, June
8. Haverkos HW, Drotman DP, Morgan WM: Kaposi's sarcoma in patients
with AIDS: Sex, transmission mode, and race. Biomed Pharmacother
9. Beral V, Peterman TA, Berkelman RL, et al: Kaposi's sarcoma
among patients with AIDS: A sexually transmitted infection? Lancet