132
L. J. Ware et al.
within this patient population, using evidence from detailed
metabolic studies to determine both the most appropriate
nutritional advice and the potential for nutritional or
pharmaceutical intervention to modulate the side effects of
these life-maintaining therapies. In using this model to
better understand effective nutrient partitioning, some
insight may be gained that will aid the management of
many diseases where an impaired and inefficient nutrient
partitioning appears to be the primary or secondary
pathology.
and the individual’s nutritional status (Nunez & Christeff,
1994). The oxidation of fatty acids is also increased in HIV
infection (Hommes et al. 1991; Mulligan et al. 1993),
although whether this increase is due to the virus, the
elevated plasma TAG concentration or increased NEFA flux
remains unclear. Plasma cholesterol (Grunfeld et al. 1992)
and phospholipid (Klein et al. 1992) concentrations tend to
be lower in HIV-infected subjects compared with healthy
controls. Polyunsaturated fatty acids (PUFA) were found to
decrease in erythrocyte membranes of AIDS patients, and a
correlation was found with both elevated malondialdehyde
concentration (the endproduct of lipid peroxidation) and a
HIV infection and metabolic control
3
CD4 count of < 50 cells/mm (Constans et al. 1995).
Hypertriacylglycerolaemia, defined as a fasting plasma
triacylglycerol (TAG) of > 2·3 mmol/l (Tikkanen, 1992),
predominantly due to increased VLDL, is prevalent in AIDS
and is positively correlated with the cytokine interferon-α
Reduced concentrations of antioxidants such as Se, vitamin
A (Sappey et al. 1994; Constans et al. 1995), glutathione
(Jahoor et al. 1999) and vitamin E (Pacht et al. 1997) have
also been found in plasma of HIV-infected patients
compared with healthy controls.
(Grunfeld & Feingold, 1991, 1992; Constans et al. 1994).
Interferon-α also shows a positive correlation with
decreased TAG clearance (Grunfeld et al. 1992) and
increased fasting hepatic synthesis of fatty acids (Hellerstein
et al. 1993). However, it is not known if interferon-α
directly induces hypertriacylglycerolaemia or is a marker
for the immune system and possibly other cytokines
The mechanism by which HIV infection induces these
alterations in lipid metabolism remains unknown. In
addition, the further elevations of plasma TAG with the
progression of HIV infection to symptomatic AIDS would
suggest a positive relationship between the extent of
viral burden and plasma TAG. However, in HIV-infected
individuals with elevated viral load effective and rapid
reduction in viraemia through administration of highly-
active antiretroviral therapy (HAART) has been shown, in
some cases, to increase plasma TAG (Carr et al. 1998) and
not to reduce lipids within the circulation, as would be
expected.
(Grunfeld & Feingold, 1991). Some research groups have
shown that a less profound hypertriacylglycerolaemia exists
in asymptomatic HIV infection in comparison with healthy
controls (Hellerstein et al. 1993). However, other
researchers state that hypertriacylglycerolaemia is only
present in the advanced symptomatic stages of infection
(AIDS), and has an inverse correlation with CD4 T-cell
count (Gomez-Sirvent et al. 1994; Christeff et al. 1995) and
a positive correlation with β2-macroglobulin and tumour
necrosis factor α (Fernandez-Miranda et al. 1998). Dietary
fish oil has been shown to alter the cytokines released by
lymphocytes during HIV infection (Bell et al. 1996),
although the implications of this change on lipid metabolism
are unknown. Given the known roles of specific fatty acids
in immune activation (Calder, 1996) and cytokine
production (Endres, 1993; Endres et al. 1989; Meydani et al.
Antiretroviral therapy and the metabolic syndrome
The development of this metabolic syndrome (termed lipo-
dystrophy due to alterations in body fat distribution) in
successfully-treated HIV patients creates a paradox. In these
patients a combination of HAART is used to effectively
reduce viral burden and allow partial immune reconstitution,
resulting in a better prognosis for the patient in terms of their
HIV disease (Gazzard et al. 1999). However, the metabolic
complications associated with chronic infection appear to be
exacerbated, resulting in much higher levels of plasma lipid,
with reduced glucose tolerance and insulin sensitivity
(Batterham et al. 2000; Carr & Cooper, 2000), in addition to
a loss of subcutaneous adipose tissue and increased visceral
adiposity in some patients (Silva et al. 1998). These
morphological and metabolic changes are reported in up to
84% of patients with long-term HAART use (Carr et al.
1999). The implications of these alterations, in conjunction
with increased life expectancy, are the increasing risk of
cardiovascular disease (Koppel, 2000; Mercie, 2000),
diabetes (Meyer et al. 1998; Carr et al. 1999) and pancrea-
titis (Sullivan et al. 1998) within this patient population.
Furthermore, patients with this change in body habitus
develop a disinclination to continue therapy, which dramati-
cally reduces life expectancy.
1
991), there may be an opportunity to modulate both these
metabolic perturbations and the immune response to chronic
infection through nutritional intervention.
In vitro studies show that HIV destroys host T-cells via a
reduced host phospholipid synthesis and an increased
neutral lipid synthesis, producing changes in host-cell
membrane permeability similar to the action of paramyx-
oviruses (Cloyd et al. 1991). However, the relative
contribution of this process to the observed elevations in
plasma TAG is unknown. In vivo studies and animal
experiments suggest that the marked increases in plasma
TAG concentration appear to be primarily due to a combi-
nation of: (a) a reduced clearance time of TAG from plasma
due to a reduction in endothelial lipoprotein lipase (LPL)
enzyme activity (Grunfeld et al. 1991); (b) an increase in
hepatic fatty acid synthesis (lipogenesis; Hellerstein et al.
1
993); (c) an increased mobilisation of free fatty acids from
Many theories have been put forward to explain how
these antiretroviral drugs could be directly responsible for
the development of this metabolic syndrome, and there
appear, currently, to be two primary schools of thought. The
first theory is that the viral protease against which the
peripheral adipocytes by lipolysis (Feingold et al. 1990;
Mulligan et al. 1993). Plasma non-esterified fatty acid
(
NEFA) concentration may be elevated or reduced
compared with controls, depending on the stage of infection