Kaposi sarcoma

Study objective: To describe the radiographic features of intrathoracic Kaposi's sarcoma in women with AIDS.

Subjects and methods: From 1987 to 1998, we identified seven women with biopsy-proven (n = 4) or autopsy-proven (n = 3) pulmonary Kaposi's sarcoma. Charts were reviewed for HIV risk factors, cutaneous and/or oropharyngeal Kaposi's sarcoma, CD4 cell count, and differential diagnosis of pulmonary disease prior to the diagnosis of pulmonary Kaposi's sarcoma. Chest radiographs (n = 6), chest CT scans (n = 3), and reports of unavailable chest radiograph (n = 1) closest to the time of diagnosis of pulmonary Kaposi's sarcoma were reviewed for the following: nodular and peribronchovascular opacities; thickened interlobular septa; pleural effusions; lymphadenopathy; and radiographic stage.

Results: Mean patient age was 33 years (range, 27 to 42 years). HIV risk factors were IV drug use (n = 2), heterosexual contact (n = 3), and both (n = 2). All patients had prior opportunistic infections. The median CD4 cell count was 18/[micro]L (mean, 63/[micro]L; range, 5 to 210/[micro]L). Cutaneous Kaposi's sarcoma was diagnosed prior to pulmonary Kaposi's sarcoma in four patients, subsequently in two patients, and not identified in one patient. Oropharyngeal Kaposi's sarcoma was diagnosed prior to pulmonary Kaposi's sarcoma in three patients. Only infection was considered in the differential diagnosis of the patients' pulmonary disease in five patients. One patient presented with acute hemoptysis and died, and one patient recently received a diagnosis of pulmonary Kaposi's sarcoma at another hospital. Chest radiographic findings were the following: nodular opacities in five of seven patients (71%); peribronchovascular opacities in six of seven patients (86%); thickened interlobular septa in two of seven patients (29%); pleural effusion in three of seven patients (43%); and lymphadenopathy in two of seven patients (29%). Five of seven patients (71%) were determined to be in radiographic stage 3, one patient in stage 1, and one patient in stage 2. CT demonstrated additional lymphadenopathy in three of three patients, thickened interlobular septa in two of three patients, and pleural effusion in one of three patients, but it did not change the staging of disease in any patient.

Conclusion: Pulmonary Kaposi's sarcoma can cause diffuse lung disease in women with AIDS. The disease is usually mistaken clinically for pulmonary infection.

(CHEST 2000; 117:410-414)

Key words: AIDS; Kaposi's sarcoma; thorax; radiology

Abbreviations: HHV-8 = human herpesvirus 8; TB = tuberculosis

An epidemic of disseminated Kaposi's sarcoma in homosexual men in 1981 was the initial impetus to investigate and describe the disease now known as AIDS. Since then, in the United States, AIDS-associated Kaposi's sarcoma has continued to be a disease that predominantly affects homosexual and bisexual men.[1] In those men who are not severely immunocompromised, Kaposi's sarcoma may remain an indolent cutaneous disease. However, in patients with advanced AIDS, Kaposi's sarcoma will often disseminate to involve the oropharynx, larynx, tracheobronchial tree, lungs, and other viscera.[1-2]

The literature describing the findings of intrathoracic Kaposi's sarcoma consists of several series of men[3-5] and includes an occasional woman.[6,7] In our inner-city population, we have encountered a number of HIV-infected women with diffuse lung disease who ultimately received a diagnosis of pulmonary Kaposi's sarcoma. The purpose of this study is to describe the radiographic findings of intrathoracic Kaposi's sarcoma in these women.

MATERIALS AND METHODS

We retrospectively identified seven HIV-infected women with pulmonary Kaposi's sarcoma between 1987 and 1998. The diagnosis was made by transbronchial biopsy in three women, by autopsy in three, and by CT-guided needle biopsy in one, All charts were reviewed for HIV risk factor, CD4 cell count, prior opportunistic infections, differential diagnosis for the patients' pulmonary disease prior to the tissue diagnosis of pulmonary Kaposi's sarcoma, and the presence of cutaneous and oropharyngeal Kaposi's sarcoma. Chest radiographs (n = 6) or reports of unavailable chest radiograph (n = 1) and chest CT scans (n = 3) were reviewed using the criteria and radiographic staging of Gruden et al[3] for the presence of nodular opacities, peribronchovascular opacities, thickened interlobular septa, pleural effusions, and hilar or mediastinal lymphadenopathy. Radiographic stages of disease are defined as follows: stage 1, isolated peribronchial cuffing; stage 2, small nodules; and stage 3, large nodules or areas of consolidation.

RESULTS

The mean patient age was 33 years (range, 27 to 42 years). HIV risk factors were heterosexual contact in three patients, IV drug use in two, and both in two. The median CD4 cell count was 18/[micro]L (range, 5 to 210/[micro]L) with a mean of 63/[micro]L. Five patients were Hispanic, one was African American, and one was African. All patients had prior opportunistic infections including the following: four instances of Mycobacterium avium complex; four instances of Pneumocystis carinii pneumonia; three instances of recurrent bacterial pneumonia; two instances of Salmonella sepsis; and one instance each of tuberculosis (TB) Mycobacterium xenopi, GI cytomegalovirus, GI cryptosporidiosis, esophageal candidiasis, toxoplasmosis of the brain, and disseminated herpes zoster. In five patients, only infection was considered in the differential diagnosis of their pulmonary disease prior to the diagnosis of Kaposi's sarcoma. The differential diagnosis in these patients included bacterial pneumonia (n = 3), P carinii pneumonia (n = 3), and M avium complex (n = 1). One patient had a recent proven diagnosis of pulmonary Kaposi's sarcoma by bronchoscopy at another institution. Her pulmonary disease was attributed to her known diagnosis. One patient presented with acute hemoptysis and died, with the diagnosis of Kaposi's sarcoma established at autopsy. Six patients had proven cutaneous Kaposi's sarcoma. The diagnosis of cutaneous Kaposi's sarcoma was made prior to the diagnosis of pulmonary Kaposi's sarcoma in four patients, subsequent to the diagnosis of pulmonary Kaposi's sarcoma in two, and was not diagnosed in one. Oropharyngeal Kaposi's sarcoma was diagnosed prior to pulmonary Kaposi's sarcoma in three patients.

The chest radiographs (Table 1) demonstrated stage 3 disease (Fig 1) in five patients, stage 1 disease in one patient, and stage 2 disease in one patient (Fig 2). In comparison, chest CT scans performed in three patients demonstrated additional peribronchovascular opacities (n = 1), thickened interlobular septa (n = 2; Fig 3), pleura] effusion (n = 1), and lymphadenopathy (n = 3). However, all three patients examined with CT had stage 3 disease demonstrated on both chest radiographs and CT scans.

[Figures 1-3 ILLUSTRATION OMITTED]

Table 1--Chest Radiographic and CT (n = 3) Findings and Radiographic Stages in Seven HIV-Infected Women With Pulmonary Kaposi's Sarcoma(*)

(*) Values given as findings for chest radiograph (CT).

([dagger]) Scale: 0 = none; 1 = 1 cm; 2 = 2 cm.

([double dagger]) Scale: 0 = none; 1 = peribronchial cuffing; 2 = coalescent perihilar.

[sections] Scale: 0 = none; 1 = [is less than] 5; 2 = many.

[parallel] Scale: 0 = none; 1 = unilateral; 2 = bilateral.

[paragraph] Scale: 1 = peribronchial cuffing; 2 = small nodules; 3 = large nodules/consolidation.

DISCUSSION

In the United States, 90 to 95% of cases of Kaposi's sarcoma occur in HIV-infected homosexual and bisexual men. Intrathoracic involvement is seen in about one third of men with cutaneous Kaposi's sarcoma and, consequently, should be strongly considered when these patients develop diffuse lung disease.[8] To our knowledge, this is the first series describing the radiographic findings of pulmonary Kaposi's sarcoma in women with AIDS. The majority of these patients were initially suspected of having an infectious etiology for their diffuse pulmonary disease. In fact, in three patients, the correct diagnosis was established only at autopsy. Two of the six women with proven cutaneous disease had that diagnosis made after the diagnosis of pulmonary Kaposi's sarcoma prompted a careful skin examination. This can be attributed to a low index of suspicion for diagnosing Kaposi's sarcoma in women with AIDS. Gender bias in the differential diagnosis of pulmonary disease in women with AIDS has been previously described. HIV-infected women with P carinii pneumonia[9,10] and with TB[11] have been reported to be both underdiagnosed and undertreated compared with men receiving the diagnosis in the same population. This also seems to be the case for pulmonary Kaposi's sarcoma.

The women in this series had chest radiographic and CT scan findings that were similar to those findings previously described for intrathoracic Kaposi's sarcoma.[3-7] Chest radiographic findings included the following: lung nodules (71%); peribronchovascular opacities (86%); thickened interlobular septa in (9.9%); pleural effusion (41%); and lymphadenopathy (9.9%). Seventy-one percent of the patients in this series had stage 3 disease (advanced intrathoracic disease) compared with 50% of patients in the series of men by Gruden et al.[3] The diagnosis of Kaposi's sarcoma at an advanced stage in women was also reported by Cooley et al,[12] who described 12 women who received a diagnosis of Kaposi's sarcoma at the Boston City AIDS Clinic over a 7-year period. Eighty-three percent of these women had advanced disease (not confined to the skin) compared with 25% of men who received a diagnosis of Kaposi's sarcoma during the same time period.

Epidemiologic evidence has suggested a sexually transmitted infectious etiology for Kaposi's sarcoma for some time, and recently the human herpesvirus 8 (HHV-8) has been described as the likely etiologic infection.[13-16] The prevalence of HHV-8 varies among different populations. It is endemic in Africa. In the United States, the prevalence of serum antibody to HHV-8 is 1% in the general blood donor population, 35% among HIV-infected homosexual men, and 4% among HIV-infected women.[17] This correlates closely with the development of Kaposi's sarcoma in HIV-infected persons. In the United States, Kaposi's sarcoma develops in 20 to 30% of HIV-infected homosexual and bisexual men and in only 1% of HIV-infected women.[17]

Heterosexual contact is the primary mode of HIV transmission worldwide.[14] National AIDS surveillance data indicates that heterosexual contact is the risk factor for HIV infection that is increasing most rapidly in the United States.[18,19] Heterosexual transmission of HIV is occurring more frequently and in disproportionate numbers among black and Hispanic women. While HHV-8 is still most prevalent in the male homosexual and bisexual population, there is a crossover into the heterosexual population. Eleven percent of AIDS patients with Kaposi's sarcoma from a recent Boston City AIDS Clinic series were women.[12] The high prevalence of Kaposi's sarcoma in these women was attributed to sexual transmission of both HIV and HHV-8 by contact with bisexual men, prostitution, or promiscuous heterosexual activity. In the present series of pulmonary Kaposi's sarcoma in seven Hispanic and black women from an inner-city population, the HIV risk factors were heterosexual contact, IV drug use, or both. We postulate that HHV-8 was sexually transmitted in all cases, as IV drug use is closely associated with sexual promiscuity. Although Kaposi's sarcoma remains predominantly a male disease at the present time, a heightened awareness of the occurrence of this disease in women may lead to diagnosis and treatment at an earlier stage.

A limitation of the present study is the small number of women comprising the study population. We believe this to be a reflection of the rarity of the diagnosis of pulmonary Kaposi's sarcoma in women with AIDS in the United States. This also may reflect the fact that bronchoscopic visualization of typical violaceous endobronchial lesions in patients with cutaneous Kaposi's sarcoma is usually considered diagnostic of intrathoracic Kaposi's sarcoma. Patients with the endobronchial lesions often do not undergo biopsy due to the risk of hemorrhage. Therefore, our strict inclusion criteria may have excluded some women with pulmonary Kaposi's sarcoma. The radiologic literature includes two series of patients with intrathoracic Kaposi's sarcoma in which one woman is described in each.[6,7] The largest clinical series, to our knowledge, of women with AIDS and cutaneous Kaposi's sarcoma, visceral Kaposi's sarcoma, or both[12] was composed of only 12 patients. Therefore, our series of seven women with AIDS and intrathoracic Kaposi's sarcoma is, to our knowledge, the largest series described in the literature to date.

In summary, we describe a group of women with AIDS and diffuse lung disease who ultimately received a diagnosis of pulmonary Kaposi's sarcoma. In most cases, despite the presence of cutaneous lesions and typical radiographic findings, the presumptive clinical diagnosis was pulmonary infection. This can be attributed to the low index of suspicion for the diagnosis of Kaposi's sarcoma in women with AIDS, which reflects a gender bias. Although uncommon, pulmonary Kaposi's sarcoma should be considered in the differential diagnosis of diffuse lung disease in women with AIDS.

ACKNOWLEDGMENT: We thank Eleanor Murphy for her assistance in manuscript preparation.

REFERENCES

[1] Friedman-Kien AE, Saltzman BR. Clinical manifestations of classical, endemic African and epidemic AIDS-associated Kaposi's sarcoma. J Am Acad Dermatol 1990; 22:1237-1250

[2] Ognibene FP, Steis RG, Macher AM. Kaposi sarcoma causing pulmonary infiltrates and respiratory failure in the acquired immunodeficiency syndrome. Ann Intern Med 1985; 102: 471-475

[3] Gruden JF, Huang L, Webb WR, et al. AIDS-related Kaposi sarcoma of the lung: radiographic findings and staging system with bronchoscopic correlation. Radiology 1995; 195:545-552

[4] Naidich DP, Tarras M, Garay SM, et al. Kaposi's sarcoma CT-radiographic correlation. Chest 1989; 96:723-728

[5] Davis SD, Henschke CI, Chamides BK, et al. Intrathoracic Kaposi sarcoma in AIDS patients: radiographic-pathologic correlation. Radiology 1987; 163:495-500

[6] Traill ZC, Miller RF, Shaw PJ. CT appearances of intrathoracic Kaposi's sarcoma in patients with AIDS. Br J Radiol 1996; 69:1104-1107

[7] Khalil AM, Carette MF, Cadranel JL, et al. Intrathoracic Kaposi's sarcoma CT findings. Chest 1995; 108:1622-1626

[8] Cadranel J, Mayaud C Intrathoracic Kaposi's sarcoma in patients with AIDS. Thorax 1995; 50:407-414

[9] Bastian LA, Sloane RJ, DeHovitz, et al. Gender differences in care for acquired immunodeficiency syndrome-related Pneumocystis carinii pneumonia. Womens Health Issues 1998; 8:45-52

[10] Horner RD, Bennett CL, Achenbach C, et al. Predictors of resource with utilization for hospitalized patients with Pneumocystis carinii pneumonia: a summary of effects from the multi-city study of quality of PCP care. J Acquir Immune Defic Syndr Hum Retrovirol 1996; 12:379-385

[11] Cegielski JP, Goetz MP, Jacobson JM, et al. Gender differences in early suspicion of tuberculosis in hospitalized, high-risk patients during 4 epidemic years 1987-1990. Infect Control Hosp Epidemiol 1997; 18:237-243

[12] Cooley TP, Hirschhorn LR, O'Keane JC. Kaposi's sarcoma in women with AIDS. AIDS 1996; 10:1221-1225

[13] Foreman KE, Friborg J, Kong W-P, et al. Propagation of a human herpesvirus from AIDS-associated Kaposi's sarcoma. N Engl J Med 1997; 336:163-171

[14] Whitby D, Howard MR, Tenant-Flowers M, et al. Detection of Kaposi sarcoma associated herpesvirus in peripheral blood of HIV-infected individuals and progression to Kaposi's sarcoma. Lancet 1995; 346:799-802

[15] Martin JN, Ganem DE, Osmond DH, et al. Sexual transmission and the natural history of human herpesvirus 8 infection. N Engl J Med 1998; 338:948-954

[16] Kedes DH, Operskalski E, Busch M, et al. The seroepidemiology of human herpesvirus 8 (Kaposi's sarcoma-associated herpes virus): distribution of infection in KS risk groups and evidence for sexual transmission. Nat Med 1996; 2:918-924

[17] Kedes DH, Ganem D, Ameli N, et al. The prevalence of serum antibody to human herpesvirus 8 (Kaposi-sarcoma-associated herpes virus) among HIV-seropositive and high-risk HIV-seronegative women. JAMA 1997; 277:478-481

[18] Neal JJ, Fleming PL, Green TA, et al. Trends in heterosexually acquired AIDS in the United States, 1988 through 1995. J Acquir Immune Defic Syndr Hum Retrovirol 1997; 14:465-474

[19] Fleming PL, Ward JW, Karon JM, et al. Decline in AIDS incidence and deaths in the USA: a signal change in the epidemic. AIDS 1998; 12(suppl A):555-561

(*) From the Department of Radiology, Albert Einstein College of Medicine, Montefiore Medical Center and Jacobi Medical Center, Bronx, NY.

Manuscript received March 12, 1999; revision accepted July 7, 1999.

Correspondence to: Linda B. Haramati, MD, FCCP, Department of Radiology, Albert Einstein College of Medicine and Montefiore Medical Center, 111 East 210th St, Bronx, NY 10467; e-mail: lharamati@aecom.yu.edu

COPYRIGHT 2000 American College of Chest Physicians
COPYRIGHT 2000 Gale Group