| |
Development of multiple lipomas during treatment with rosiglitazone in a patient with HIV-associated lipoatrophy
|
| |
| |
AIDS: Volume 18(12) 20 August 2004
Mafong, Derek Da,d,f; Lee, Grace Aa,d; Yu, Siegridb,e; Tien, Phyllisa,c,d; Mauro, Theodorab,e; Grunfeld, Carla,d
aMetabolism and Endocrine Sections, bDermatology Service, and cInfectious Diseases Section, San Francisco Department of Veterans Affairs Medical Center, San Francisco, CA, USA; Departments of dMedicine, and eDermatology, University of California, San Francisco, CA, USA; and fDepartment of Medicine, University Hospitals of Cleveland, Cleveland, OH, USA.
The first two authors contributed equally.
HIV-associated lipoatrophy is common in individuals on highly active antiretroviral therapy. The syndrome is characterized by varying degrees of lipoatrophy in the subcutaneous fat of the face and limbs. It is often associated with insulin resistance and dyslipidemia. Several therapies to improve some of these metabolic abnormalities have been explored, although no single treatment has yet been substantiated. Thiazolidinediones are insulin-sensitizing agents that decrease visceral fat and increase subcutaneous fat. They are peroxisome proliferator-activated receptor gamma (PPARy) agonists that have been shown to promote fat cell differentiation and inhibit the toxic effects of HIV protease inhibitors on adipogenesis in vitro. As a consequence, thiazolidinediones have been used as a potential treatment for patients with lipoatrophy in the absence or presence of HIV. Four studies have reported small improvements in body fat distribution or metabolic parameters in HIV-infected patients treated with thiazolidinediones. Here we report the first case of reversible thiazolidinedione-induced lipomatosis in a patient with HIV-associated lipoatrophy.
The patient was a Caucasian man with a history of HIV infection diagnosed in 1992 at the age of 58 years, who had had type 2 diabetes mellitus since 1993, which was well controlled with insulin. He was initially treated with zidovudine in 1993, which was discontinued after 3 months because of fatigue. Over the next 4 years without antiretroviral therapy, his absolute CD4 cell count gradually decreased from 510 cells/mm3 to a nadir of 266 cells/mm3. Because of this decline in his CD4 cell count, highly active antiretroviral therapy with stavudine, lamivudine, and nelfinavir, was initiated in 1997 with an excellent immunological and virological response. His CD4 cell count rose to 547 cells/mm3, with an undetectable HIV viral load. Lipoatrophy, with no visible facial fat and extremities with prominent veins, was first noticed in February 2001.
In March 2002, rosiglitazone was started as an insulin-sensitizer and secondarily to determine whether it would increase peripheral fat. The patient weighed 214 lb immediately before starting rosiglitazone. Over the next 3 months, he developed several dozen lipomas bilaterally in his arms and upper thighs, measuring 1-4 cm in diameter. His weight remained at 214 lb. A biopsy of one of these lesions on the left upper arm revealed a well-circumscribed tumor of normal-appearing fat cells, located beneath normal-appearing skin. Normal epidermis and dermis overlying the lipoma and lipoma tissue were labeled with an anti-PPARy antibody (sc-7196, Santa Cruz Biotechnology, Santa Cruz, CA, USA), subsequent anti-rabbit fluorescein-isothiocyanate-labeled secondary antibody, and nuclear-counterstained with propidium iodide. The lesion was a lipoma, which stained for PPARy. On review of his medical history at this time, he reported a history of four to five lipomas in his teenage years. Rosiglitazone was discontinued. Within 3 months, all but five lipomas resolved completely (see Fig. 1d). The patient reported that these remaining lipomas were different from the lipomas he had as a teenager. After stopping rosiglitazone, his weight ranged from 207 to 222 lb over the next 13 months.
Studies on the effects of thiazolidinediones in lipodystrophy have focused on quantifying insulin resistance and body fat distribution in patients with HIV-associated lipodystrophy. Most studies have been pilot studies showing small increases in measured subcutaneous body fat and small decreases in visceral fat, often with no visible change in fat. The adverse effects of thiazolidinediones in this population are unknown. Given the promotion of fat cell differentiation, increased lipomatosis is a possible side-effect of thiazolidinediones in HIV-infected patients with lipoatrophy. The finding of PPARy in the lipoma of this patient and the reversibility of his lipomatosis on the discontinuation of rosiglitazone support this linkage. Because this patient had no visible gain in fat in non-lipomatous areas, but an increase in lipomas after starting rosiglitazone, the proliferation of lipomas raises the possibility of a differential regulation of lipomas and fat. However, a caveat is that body fat measurements were not obtained so that small changes in non-lipomatous fat could have been missed. Given that lipomas in the general population are not uncommon, thiazolidinedione therapy should be approached with caution in patients with HIV-associated lipoatrophy.
|
|
| |
| |
|
|
|
