Objectives: Body weight is regulated by the balance between energy int
ake and energy expenditure, but the influence of HIV infection on ener
gy balance has not been fully examined. The main objectives of this st
udy were (1) to assess the effect of HIV on energy balance, (2) to exa
mine the relationship of parameters of immunodeficiency to energy bala
nce, and (3) to examine the interrelationship of different components
of energy balance in asymptomatic HIV-seropositive men. Design: A cros
s-sectional study of nutrition and metabolism in asymptomatic HIV-sero
positive men. Methods: Components of energy balance were examined in 1
04 asymptomatic HIV-seropositive men (CD4 count 4-482x10(6)/1) and 57
age-matched HIV-seronegative male controls. Energy and protein intake
were measured using 5-day diaries, and small bowel absorption and perm
eability was assessed using four sugar probes. Resting energy expendit
ure was calculated from indirect calorimetry and nitrogen loss estimat
ed from 24 h urine collection. Four methods were used to assess the ef
fect of HIV infection on body composition (anthropometry, dual energy
X-ray absorptiometry, bioelectrical impedance and 24 h urine creatinin
e). Results: Resting energy expenditure per kilogram of fat-free mass
was raised (P < 0.0001), fat mass was decreased (P = 0.001), fat-free
mass was increased (P = 0.05), energy intake was higher (P = 0.05), ab
sorption of L-rhamnose (P = 0.01) and 3-O-methyl-D-glucose was decreas
ed (P = 0.003), and small bower permeability was increased (P < 0.0001
) in HIV-seropositive men compared with HIV-seronegative controls. HIV
-seropositive subjects with a CD4 count less than 100x10(6)/l had decr
eased absorption of L-rhamnose (P < 0.05), D-xylose (P < 0.05) and 3-O
-methyl-D-glucose (P < 0.05) compared with HIV-seropositive subjects a
t higher CD4 counts, and had a similar resting energy expenditure to H
IV-seronegative controls. Protein intake, carbohydrate, fat and protei
n oxidation, 24 h nitrogen excretion and appendicular muscle mass were
similar in HIV-seropositive men and controls. Conclusion: HIV infecti
on exerts a direct effect on parameters of energy balance that varies
with the severity of immunosuppression.