An Unusual Case of Chylothorax in an HIV-infected Patient with Kaposi's Sarcoma

Mitchell L. Margolis, MD, Director, Clinical Pulmonary Medicine, Pulmonary Section, Philadelphia VA Medical Center, Clinical Associate Professor, Pulmonary, Allergy, Critical Care, and Sleep Division, University of Pennsylvania; Eileen B. MacDonald, MD, Instructor, Department of Medicine, University of Pennsylvania; Phillip Green, MD, Attending Physician, Infectious Diseases Section, Philadelphia VA Medical Center, Clinical Assistant Professor, Division of Infectious Diseases, University of Pennsylvania; Gregg Y. Lipschik, MD, Director, Medical Intensive Care Unit, Pulmonary Section, Philadelphia VA Medical Center, Clinical Associate Professor, Pulmonary, Allergy, Critical Care, and Sleep Division, University of Pennsylvania, Philadelphia, Pa; Guest Editor: H. Ralph Schumacher, Jr, MD, Professor of Medicine, University of Pennsylvania School of Medicine, Philadelphia

The presence of chylothorax in HIV-infected patients with Kaposi's sarcoma (KS) has been a devastating complication marked by progressive weight loss, weakness, protein loss, lymphopenia, and further immune compromise. In previous reports, its occurrence usually signified widespread disease and portended an extremely poor prognosis.^1-9 In contrast, our experience with the patient described here was atypical, with its favorable outcome, which prompted a brief review of chylothorax and the limited medical literature devoted to this condition in patients with AIDS and KS.

Case Presentation

A 44-year-old HIV-infected man with a history of intravenous (IV) drug use and homosexuality was hospitalized because of shortness of breath over the past several months. He had stable KS involving the skin and soft tissues of the anterior thighs for 7 years. He had begun highly active antiretroviral therapy (HAART) 4 years before the current illness. His CD4⁺ cell count was 280/?L, and his viral load was less than 50 copies/mL. His medical history included a complex illness 9 months ago, during which he underwent resection and debridement of a right sternoclavicular joint abscess, requiring substantial mediastinal dissection. During the perioperative period, several vascular access devices were inserted into the left subclavian vein for long-term IV antibiotic therapy.

Physical examination revealed no change in the superficial KS lesions, but dullness to percussion and absent breath sounds were detected at the right-lung base. Chest x-ray showed a moderate-sized pleural effusion on the right (Figure), which was not present at his previous admission. A chest computed tomography (CT) scan confirmed a right pleural effusion but no pulmonary lesions or mediastinal or hilar adenopathy. Thoracentesis revealed 1.5 liters of milkywhite fluid with the biochemical profile of a chylothorax, including a triglyceride level of 4094 mg/dL. Fiberoptic bronchoscopy was unremarkable.

Dyspnea and pleural fluid recurred during the next few weeks, necessitating 2 therapeutic thoracenteses. Lymphangiographic coil embolization of the thoracic duct was attempted, but cannulation of the cisterna chyli was unsuccessful. Eventually, right thoracoscopic pleurodesis was successful. The patient remains asymptomatic and free of pleural fluid reaccumulation 2 years after the procedure.

Discussion

Our patient presents an unusual case of KS associated with chylothorax that appears not to have resulted from overwhelming KS infiltration of the lung and thoracic lymph nodes, which has usually been the case in previous patients reported in the literature.

The term "chylothorax" refers to the accumulation of chyle in the pleural space. Chyle comprises lymph of intestinal origin?a milky, opalescent fluid rich in lymphocytes, protein, triglycerides, and chylomicrons. Normally, chyle is conducted from intestinal lacteals and lymphatics to the cisterna chyli, a triangular dilatation overlying the anterior surface of the second lumbar vertebrae. From there, chyle is transmitted via the thoracic duct as it ascends through the posterior mediastinum, eventually draining into the left subclavian vein.

For chyle to accumulate in the pleural space, there must be some disruption of flow in the thoracic duct. This results in a mediastinal collection of chyle, which in turn ruptures, thereby gaining access to the pleural space. The causes of chylothorax have been classified into 4 groups: tumor (especially lymphoma); trauma (usually surgical); idiopathic; and miscellaneous, including such diverse entities as thrombosis of the superior vena cava or subclavian vein, cirrhosis, and rare conditions, such as pulmonary lymphangiomyomatosis.¹⁰

The most common symptom of chylothorax is dyspnea, which usually results from compression of the ipsilateral lung by a sizable chyle collection. The most common complications are malnutrition and/or immune compromise, which are caused by lymphocyte, protein, and fat depletion, especially when chest tube drainage or repeated thoracenteses are attempted.

The diagnosis is suggested by the withdrawal of milky-white fluid from the pleural space during thoracentesis. However, this finding can also be seen in association with empyema or pseudochylothorax, a rare sequela of chronic rheumatoid or tuberculous effusions in which cholesterol crystals, not chyle, are responsible for the turbidity. Also, the characteristic milky appearance may be absent even in the presence of a true chylothorax. Chylothorax in malnourished patients with AIDS, for example, may lack the typical milky appearance because of poor oral fat intake and depletion of chylomicrons.^1,5

Thus, definitive diagnosis of chylothorax depends on the measurement of pleural fluid lipid levels, specifically a pleural fluid triglyceride level of more than 110 mg/dL and a pleural fluid/serum cholesterol ratio of less than 1.¹⁰ An important corollary is that a pleural triglyceride level measurement should be ordered when chylothorax is suspected, even in the absence of the characteristic gross appearance.

Initial treatment generally consists of dietary manipulation, such as switching to medium-chain triglycerides, as well as bowel rest to lessen chyle formation and thoracentesis to relieve dyspnea. If large amounts of chyle continue to form, more definitive treatment can be offered, such as thoracic duct ligation, pleurodesis, or pleuroperitoneal shunting.

In HIV-infected patients with KS, chylothorax is a rare but recognized complication.^1-9 We found only 10 previously published case reports in the medical literature (Table). Some of the reports lack specific critical details, but several general observations can still be made. KS-associated chylothorax was described in young men with preexisting cutaneous KS who then developed dyspnea, cough, and/or fever. The HIV infection risk factor, when reported, was homosexuality or bisexuality. Chest x-rays showed sizable pleural collections, usually bilateral or left-sided. The fluid was described as milky or turbid, with a high triglyceride level. Most cases had evidence of lymphatic obstruction via KS involvement of the mediastinal nodes and/or prominent pulmonary KS. Treatment included palliative measures involving pleural drainage, as well as attempts to prevent recurrence via pleural sclerosis, dietary manipulation, fluid shunting, and/or chemotherapy directed at KS. In general, treatment was unsuccessful, and patients in whom the outcome was noted died within a few weeks to months.

Subsequent to most of these reports, remarkable progress has been made in the understanding and treatment of KS. A specific herpesvirus has been implicated in the pathogenesis,¹¹ and striking reductions in incidence and improvements in survival have been noted in patients receiving HAART.^12,13 In fact, the only patient, other than ours, who had KS and chylothorax with prolonged survival was also managed with HAART.⁹

Though the exact pathogenesis of chylothorax in our patient is unclear, his case stands in marked contrast to the published experience with this disease entity. As noted, our patient had no evidence of pulmonary or mediastinal KS, responded well to pleurodesis, and is alive more than 2 years after presentation with chylothorax.

Although we cannot totally exclude subradiographic KS-related lymphatic obstruction, we believe that at least 2 other events in our patient's history could have caused the chylothorax?surgical trauma or subclavian vein thrombosis. Surgical trauma is an important cause of chylothorax,¹⁴ and our patient had an extensive mediastinal dissection 9 months before the initial detection of pleural fluid accumulation. The dissection was necessary to adequately debride a large abscess that extended from an infected right sternoclavicular joint posteriorly into the mediastinum. Since the anatomy of the thoracic duct is notoriously variable, and the vessel itself is difficult to localize intraoperatively, the duct or its tributaries may have been violated during the procedure, thereby allowing chyle to leak into the pleural space. However, the interval between the putative operative trauma and detection of chylothorax in our patient was unusually long. The interval between thoracic duct injury and the appearance of chylothorax is usually only a few days, with exceptional cases detected as long as 60 days after surgery¹¹ or 11 weeks after blunt trauma.¹⁵

In addition, our patient had several cannulations of his left subclavian vein for vascular access in the perioperative and postoperative periods. Such instrumentation can induce subclavian vein thrombosis, which has rarely been cited as a cause of chylothorax.¹⁶ The mechanism presumably involves occlusion of the orifice of the thoracic duct with obstruction of the flow of chyle into the venous system, back-pressure in communicating vessels, and leakage of chyle from pleural lymphatics into the pleural space, often on the right side.^16,17 Because venography was not done on our patient, the possibility remains that subclavian vein thrombosis played a role in initiating the chylothorax.

Based on these considerations, we believe that our patient's chylothorax?which must ultimately be labeled "idiopathic," the third most common cause¹⁰?likely was coincidental to his KS rather than a direct result of it.

Other potential causes of chylothorax in HIV-infected patients include lymphoma and tuberculosis,¹⁸ but chylothorax appears to be extremely rare in this population, and KS remains the leading concern in the differential diagnosis. Chylothorax is also rare in patients with KS but without HIV infection, who seldom manifest disseminated disease. We found only a single case of KS-related chylothorax in an HIV-negative patient reported in the past 40 years.¹⁹

Acknowledgment

The authors thank A. Ross Hill, MD, for his translation of reference 8 from the original French.

NY State J Med

1. Schulman LL, Grimes MM. Metastatic Kaposi's sarcoma and bilateral chylothorax. . 1986;86:205-206.

Chest

2. Pandya K, Lal C, Tuchschmidt J, et al. Bilateral chylothorax with pulmonary Kaposi's sarcoma. . 1988;94:1316-1317.

Chest

3. O'Brien RF, Cohn DL. Serosanguineous pleural effusions in AIDSassociated Kaposi's sarcoma. . 1989;96:460-466.

South Med J

4. Judson MA, Postic B. Chylothorax in a patient with AIDS and Kaposi's sarcoma. . 1990;83:322-324.

West J Med.

5. Pennington DW, Warnock ML, Stulbarg MS. Chylothorax and respiratory failure in Kaposi's sarcoma. 1990;152:421-422.

South Med

J

6. Priest ER, Weiss R. Chylothorax with Kaposi's sarcoma. . 1991;84:806-807.

Klin

Wochenschr

7. Bogner JR, Gross M, Zietz C, et al. Chylothorax as fatal complication in fulminating Kaposi's sarcoma in a patient with AIDS. . 1991;69:134.

Presse Med

8. Vandenbos F, Barel R, Abbyad R, et al. Chylothorax as a complication of Kaposi's sarcoma. . 1998;27:1218. [in French]

AIDS

9. Maradona, JA, Carton JA, Asensi V, et al. AIDS-related Kaposi's sarcoma with chylothorax and pericardial involvement satisfactorily treated with liposomal doxorubicin. . 2002;16:806.

Pleural Diseases

10. Light RW. . 3rd ed. Baltimore, Md: Williams & Wilkins; 1995:284-298.

J

Acquir Immune Defic Syndr

11. Gill J, Bourboulia D, Wilkinson J, et al. Prospective study of the effects of antiretroviral therapy on Kaposi's sarcoma-associated herpesvirus infection in patients with and without Kaposi's sarcoma. . 2002;31:384-390.

Int J Cancer

12. Tam HK, Zhang ZF, Jacobson LP, et al. Effect of highly active antiretroviral therapy on survival among HIV-infected men with Kaposi's sarcoma or non-Hodgkin's lymphoma. . 2002;98:916-922.

Int J Cancer

13. Carrieri MP, Pradier C, Piselli P, et al. Reduced incidence of Kaposi's sarcoma and of systemic non-Hodgkin's lymphoma in HIV-infected individuals treated with highly active antiretroviral therapy. . 2003;103:142-144.

J Surg Oncol

14. Sieczka EM, Harvey JC. Early thoracic duct ligation for postoperative chylothorax. . 1996;61:56-60.

Thorac Cardiovasc

Surg

15. Milano S, Maroldi R, Vezzoli G, et al. Chylothorax after blunt chest trauma: an unusual case with a long latent period. . 1994;42:187-190.

Am J Clin Oncol

16. Van Veldhuizen PJ, Taylor S. Chylothorax: a complication of left subclavian vein thrombosis. . 1996;19:99-101.

Klin Wochenschr.

17. Smedts F, Kubat K, Chande H. Chylopericardium and chylothorax resulting from a catheter to the left subclavian vein: an autopsy report. 1989;67:1214-1217.

Arch Intern Med.

18. Singh S, Girod JP, Ghobrial MW. Chylothorax as a complication of tuberculosis in the setting of the human immunodeficiency virus infection. 2001;161:2621.

Br J Dermatol.

19. Fife KM, Talbot DC, Mortimer P, et al. Chylous ascites in Kaposi's sarcoma: a case report. 1992;126:378-379.

Commentary

Pablo Tebas, MD, Associate Professor of Medicine, University of Pennsylvania, Philadelphia, Pa

Pneumocystis carinii

In 1981, clinicians in Los Angeles and other parts of the United States started to see an unusual cluster of homosexual men with 2 infrequent diseases, pneumonia¹ and Kaposi's sarcoma (KS).²Anew, devastating epidemic had just been recognized. Since then, we have learned more about HIV infection and AIDS than most other diseases in the history of humankind. In 3 years, the causative agent was identified and, in 15 years, thanks to the discovery of combination highly active antiretroviral therapy (HAART), what once was a fatal disease has become a manageable disease for patients who live in countries that provide access to these life-saving medications.³

HIV infection or AIDS was, and still is, a stigmatizing disease. In the 1980s, having cutaneous KS was the equivalent of carrying a sign that said, "I have AIDS." It is difficult for young physicians who did not live those days to realize how devastating it was, both physically and psychologically, for a patient to have this complication. Things have changed for the better, but physical stigma still exist for many patients. Indeed, currently the most stigmatizing physical finding for HIV-infected patients is related to the treatment of this disease rather than the disease itself, namely, the lipoatrophy associated with antiretroviral therapy.

Mycobacterium avium-intracellulare

Dr Margolis and colleagues describe a patient with an unusual complication of his HIV-associated KS, chylothorax. Everything seems to fit, and, although unusual, it appears that his chylothorax might be secondary to his KS. However, HAART has taught us that things are not that simple anymore. Successful antiretroviral treatment results in immune reconstitution. Infections that were previously impossible or difficult to treat, such as cryptosporidiosis, azole-resistant thrush, cytomegalovirus retinitis, , or progressive multifocal leukoencephalopathy, have changed their naturally inexorable courses. KS has not been exempted from this change, and HAART has simultaneously decreased its perceived incidence and improved its prognosis.⁴ However, this has not been the result of a decreased transmission of KS herpesvirus (also known as herpesvirus type 8), the causative agent of KS, but probably the result of the improved immune system of coinfected patients.⁵

The patient described in this case has a well-controlled HIV RNA viral load and an elevated CD4⁺ cell count. We should take a second look at our assumptions. Maybe what looks clear is not so clear. That is what Dr Margolis and colleagues want us to do. They make a strong case that the patient's problem is likely related to a previous surgery and not to KS or HIV infection. As they try to convince us, they do a superb review of the causes of chylothorax in HIV infection. And as often happens in clinical medicine, we cannot be absolutely certain that the new explanation is the truth, but it certainly makes more sense than the alternatives.

The main lesson from this case is that patients with well-controlled HIV infection can have other complications that are completely unrelated to their disease. It is necessary to see patients in their clinical context, specifically their degree of immune suppression if they have not started therapy, and their degree of immune reconstitution after starting therapy.

Remember to keep an open mind and consider all possibilities (HIV related or not) when you see the next unusual complication in an HIV-infected patient. Today, it is more true than ever that AIDS and HIV infection have replaced syphilis in the old aphorism of Sir William Osler: "Know syphilis in all its manifestations and relations, and all other things clinical will be added unto you."⁶

Pneumocystis carinii

N Engl J Med

1. Gottlieb MS, Schroff R, Schanker HM, et al. pneumonia and mucosal candidiasis in previously healthy homosexual men: evidence of a new acquired cellular immunodeficiency. . 1981;305:1425-1431.

Ann Intern Med

2. Fauci AS. The syndrome of Kaposi's sarcoma and opportunistic infections: an epidemiologically restricted disorder of immunoregulation. . 1982;96:777-779.

N Engl J Med

3. Palella FJ Jr, Delaney KM, Moorman AC, et al, for the HIV Outpatient Study Investigators. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. . 1998;338:853-860.

J Acquir Immune Defic Syndr

4. Jones JL, Hanson DL, Dworkin MS, et al. Effect of antiretroviral therapy on recent trends in selected cancers among HIV-infected persons. Adult/Adolescent Spectrum of HIV Disease Project Group. . 1999;21(suppl 1):S11-S17.

JAMA

5. Osmond DH, Buchbinder S, Cheng A, et al. Prevalence of Kaposi's sarcoma-associated herpesvirus infection in homosexual men at beginning of and during the HIV epidemic. . 2002;287:221-225.

Sir William Osler: Aphorisms from His Bedside Teachings

and Writings.

6. Bean RB. Collected by Robert Bennett Bean, M.D. (1874-1944). Springfield, Ill: Charles C. Thomas, 1968.