Send to

Choose Destination
Lancet. 1998 Mar 21;351(9106):867-70.

"Buffalo hump" in men with HIV-1 infection.

Author information

Division of Endocrinology, San Francisco General Hospital, CA 94110, USA.



Enlargement of the dorsocervical fat pad ("buffalo hump") has been reported in numerous HIV-1-infected patients. Some investigators have speculated that this finding is associated with protease-inhibitor treatment.


Between June, 1995, and October, 1997, we studied eight HIV-1-infected men who had developed a buffalo hump while otherwise stable on antiretroviral therapy. Measurement of 24 h urinary free cortisol excretion and an overnight low-dose dexamethasone suppression test were done to screen for Cushing's syndrome. In one patient, plasma cortisol concentrations were measured every 4 h for 24 h to assess the circadian rhythm of cortisol. Results of total and regional body-composition analysis by dual-energy X-ray absorptiometry, and glucose, cholesterol, triglyceride, and cortisol concentrations were compared with those obtained in a control population of 15 HIV-1-positive men whose age, body-mass index (BMI), and CD4-lymphocyte count were within the range of values in the eight study patients.


The eight patients with a buffalo hump were clinically stable on various antiretroviral regimens, four of which included a protease inhibitor. No other signs of Cushing's syndrome were observed, and plasma cortisol values did not differ significantly from those of controls. 24 h urinary free cortisol excretion was normal in seven patients and slightly raised in one (248 nmoles). In this patient, a repeat 24 h urinary free cortisol was 175 nmoles and plasma cortisol concentrations over 24 h showed a normal circadian pattern (nadir 83 nmol/L at 2400 h). All eight patients had normal suppression of cortisol values after dexamethasone 1 mg (plasma cortisol less than 83 nmol/L). When compared with HIV-1-positive controls, men with a buffalo hump had a significantly greater proportion of fat in the trunk region, suggesting central fat accumulation. Triglyceride but not cholesterol values were higher in the patients than in controls but this difference was not significant. Fasting glucose values did not differ significantly.


The development of a buffalo hump cannot be attributed to hypercortisolism in these eight men. Furthermore, its occurrence is not unique to patients on protease inhibitors. Although the mechanism for dorsocervical fat accumulation is unclear, we speculate that regional abnormalities in lipogenesis and lipolysis occur, possibly influenced by the hormonal and metabolic changes seen with HIV-1 infection and its treatment.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center