Issue # 46: Nutrition and HIV
Study one
124 HIV infected adults
attending an AIDS outpatient clinic in France had
nutritional assessment
undertaken, based on anthropometry, impedance
measurement and plasma albumin
and pre-albumin assay.
Result:
Normal nutrition 62%
Moderate malnutrition 13%
Intermediate malnutrition 17%
Severe malnutrition
8%
Ref: Eur J Clin Nutr 1997;51(9): 637 - 640
Study two
125 HIV seropositive drug-using
adults from Miami had blood levels of
various nutrients measured, and
were then followed for 3.5 years. Outcome
measure was death from
HIV-related causes.
Relative risk of death RR p<
------------------------------------------------------------------------
Low initial levels of:
Pre-albumin 4.01 0.007
Vitamin A 3.23 0.03
Vitamin B12 8.33 0.009
Zinc 2.29 0.04
Selenium 19.9 0.0001
These relationships were
independent of CD4 counts either at baseline or
over the follow-up period.
However, when adjusted for all variables known
to affect HIV-related
mortality, only selenium status remained
independently significant.
Ref: J Acquir Immune Defic
Syndr Hum Retrovirol 1997;15:370-4
Study three
64 HIV-infected outpatients
with CD4 counts >100 m/l were randomised to
receive either a daily 600 kcal
oral nutritional supplement with vitamins,
trace elements and minerals, or
the same supplement with additional
arginine and omega-3 fatty acids.
Results: There was an increase
in body weight in both groups, but
enrichment with arginine and
omega fatty acids did not improve
immunological parameters.
Ref: AIDS 1998;12:53-63
Studies four/five
310 HIV infected men had serum
vitamin A, E and B12 assayed in the
Baltimore-Washington area of
the USA.
Outcomes were measured over a 9
year follow-up:
- first AIDS diagnosis
- CD4+ cell decline (< 200m/l)
- mortality (vitamin A and E
only) only.
Results:
Vitamin B12: Subjects with low
serum vitamin B12 (< 120 pmol/L) had
significantly shorter AIDS-free
time than those with adequate vitamin B12
(4 vs. 8 years p = 0.004).
Vitamin E: There was a
significant association between higher vitamin E
levels and slower progression to
AIDS (highest quartile vs remaining
subjects RR=0.67, 95% CI
0.45-0.98). Similar trends were seen in relation
to mortality, but not with CD4
count. There was also a relationship between
taking vitamin E-containing
supplements and serum vitamin E levels.
Vitamin A: There were no such
clear-cut relationships, either between
supplement use and vitamin A
levels, nor between serum vitamin A and
progression of AIDS. However,
there were also no abnormally low serum
vitamin A levels found.
Ref: AIDS 1997;11:613-20; J Nutr 1997;127:345-51
COMMENT
These studies together express
nicely the intrigue and the frustration of
research into HIV nutrition.
On the one hand, the first
study confirms what is already well established
- that generalised malnutrition of mild to severe degree is
commonplace in
HIV infected patients. The second study reaffirms
dramatically what is
also well known, that
malnutrition in HIV is associated with worse clinical
outcome.
Unfortunately the third study,
like many before it, reminds us that it is
not easy to reverse this
general malnutrition, let alone to demonstrate
improvements in clinical status
from nutritional treatment.
Various supplements have been
tried to boost energy, protein and fatty acid
intakes, and whilst these approaches
have generally been able to partially
reverse weight loss, studies
showing impacts on immune function or clinical
progression are much harder to
find.
There are many possible reasons
for this. Poverty and concurrent illness
makes compliance with dietary
change difficult in many cases. The intrinsic
effects of AIDS-related
catabolism seem to be particularly resistant to
therapy, as they are in cancer.
Even more difficult to counter are the
effects of the commonly present
chronic diarrhoea and gut infections that
mark HIV progression.
The remaining two studies offer
fascinating glimpses into the possible
benefits that could result from
supplementation with specific nutrients.
Selenium status was strongly
associated with mortality over 3.5 years in
the second study, whilst serum
vitamin E and B12 appear to be linked with
progression to AIDS in the last
two studies.
These are of course only
epidemiological associations, and the need for
randomised, placebo-controlled
intervention trials is clear.
Perhaps the most surprising
thing to the editors of these Updates, however,
is just how hard it is to find
such randomised, controlled trials - after
more than a decade of awareness
of the importance of nutrition to HIV
patients.
To illustrate our point, we
entered the terms "HIV", "nutrition" and
"controlled trial"
into a Medline search for the last 5 years. We came up
with only 6 relevant trials.
Perhaps we missed something, but even so this
is clearly an area ripe for
further development.
trials.
Return to index
Issue # 47: Nutritional assessment
Study one: ECG changes
Twenty malnourished Venezualan
children and twenty control children were
assessed by electrocardiography
(ECG) and other conventional measures of
nutritional status, including
blood count, serum protein and electrolytes.
Malnourished children had lower
plasma protein and electrolyte
concentrations than the
controls. Their haemoglobin and hematocrit levels
were also lower.
There were significant
differences in their ECG patterns, particularly in
relation to the QT interval.
The malnourished children had more
abnormalities (flattened or
inverted T waves, U waves) and there was
greater corrected QT-interval
dispersion, accentuated in the precordial leads.
Ref: Clin Cardiol 1998;21:201-5
Study two: hand grip strength
Hand grip strength is an easily
measured and convenient way of assessing
skeletal muscle mass in
surgical patients, according to Chinese researchers.
The surgeons from Beijing
assessed hand grip strength in 127 patients with
oral and maxillofacial cancer.
Results correlated well and significantly
with more conventional measures
of muscle mass (mid-arm muscle
circumference and
creatinine-height index).
There was also a significant
relationship between hand grip strength and
post-operative complications.
Those whose initial hand grip strength was
less than 85% of the control
value developed significantly more
postoperative complications
than those in whom it at least 85% (48%
complications vs 18%, p =
0.004).
Ref: Br J Oral Maxillofac Surg
1996;34:325-7
Study three: body impedance analysis
Electrical impedance analysis
is a well established body anthropometrical
method, but what is its
usefulness compared with four-site skinfold
thickness? This question was
addressed in a renal dialysis setting in a
recently published Dutch study.
In renal dialysis patients it
is important to be able to monitor hydration
and lean body mass on an
ongoing basis. In this study, these variables were
assessed in twenty stable
dialysis patients before and after dialysis using
both four-site skinfold
anthropometry and bioelectrical impedance analysis.
Measured against objective
weight loss, both methods had similar correlations:
Four-site anthropometry: r = 0.75, p <0.005
Impedance analysis: r = 0.69, p < 0.005.
Similar correlations were found
for hydration status. There was a very
close correlation overall
between results from the two methods (r = 0.93, p
< 0.005), although four-site
anthropometry was less affected by changes in
fluid status.
The authors concluded that,
since the two results were essentially
equivalent, the impedance
method was preferable as it was less dependant on the interpretation of the
person measuring it.
Ref: Clin Nephrol 1998; 49:180-5
COMMENT
Studies such as these help us
are useful in showing that there are some
quite simple and useful ways of
assessing nutritional status which are well
within the scope of outpatient
medicine.
It is not surprising that
something so sensitive as the repolarisation of
heart muscle would be sensitive
to changes in nutritional status, being so
dependant as it is on
electrolyte balance and efficient operation of
energy-dependant membrane
pumps.
Lowered hand grip strength is a
well known sign of decreased muscle mass.
This study shows that, by using
a simple piece of low-tech equipment,
attending physicians can get an
assessment of definite prognostic value.
Although neither ECG analysis
nor hand grip strength could yet claim the
status of established clinical
tests of nutritional state, these results
suggest that they could be
developed into useful adjuncts.
The other thing that these
studies help us with is in recognising that
nutritional status assessment
needs to be tailored to the individual
clinical situation. What is
most important to measure in a cancer patient
in the surgical ward is not the
same as in a renal patient on dialysis.
The conventional approach to
teaching clinicians to assess nutritional
status emphasises general
measures, such as weight/BMI, skinfold thickness
and serum proteins. This is
good in that it focuses on the larger picture
and treats nutritional status
holistically. But there is also a value in
teaching doctors to look at
more specific parameters in particular clinical
situations.
.
trials.
Return to index
Issue # 48: Phytoestrogens and the heart
Study one
A higher intake of soy products
is associated with decreased serum total
cholesterol levels, according
to a Japanese study of nearly five thousand
adults.
A semi-quantitative food
frequency questionnaire was used to assess usual
dietary intake in 1,242 men and
3,596 women during an annual check-up.
Fasting blood cholesterol was
measured.
In men, there was a significant
correlation between increasing intake of
soy products and total
cholesterol concentration (p = 0.0001, after
controlling for possible
confounders)
A similar negative correlation
was seen in women (p=0.00001, after
controlling for similar
confounders, and also menopausal status).
Ref: J Nutr 1988; 128:209-213
Study two
The effects of soybean protein
isolate in relation to lipid levels after a
meal were examined in 11 men
with normal lipid profiles. Oral fat load
tests were performed before and
3 weeks of either soy protein isolate or
casein supplementation (20
g/d).
Neither the soy protein nor the
casein supplement affected fasting plasma
lipids or apolipoproteins, but
the curve of the level of remnant-like
particles of cholesterol after
the meal was significantly smaller in area
(p < 0.05) after the soy supplemented
arm of the study compared with the
casein supplementation.
J Nutr Sci Vitaminol (Tokyo)
1998;44:113-27
Study three
An double blind,
placebo-controlled intervention trial in Australia failed
to demonstrate any effect of
isoflavonoid supplementation on lipid levels.
The trial involved 46 middle
aged men (35-69 years) and 13 postmenopausal
women with initial cholesterol
readings of average < 5.5 mmol/L . They were
given a tablet containing 55 mg
of isoflavonoids or placebo for 8 weeks.
The tablet, which was derived
from red clover extract and contained a
predominance of genistein.
Subjects continued their normal diet. in all
other respects.
There was no signficant
difference between supplemented and placebo groups
after intervention in relation
to any lipid parameter measured, including
total, LDL, HDL and HDL
subclass cholesterol, triglycerides and lipoprotein
A. There was also no
significant correlation between changes in urinary
isoflavonoids and those in
serum lipids.
Ref: J Nutr 1998;128:728-732
Study four
Isoflavones enhance arterial
elasticity, according to results of another
Australian study. A group of 21 menopausal and perimenopausal
women were
given either isoflavone
(80mg/day) or placebo in a crossover paradigm over
five to ten week periods.
There was a 26% improvement in
systemic arterial compliance (p < 0.001) in
the isoflavone supplemented
women compared with placebo. There was no
impact on arterial pressure and
plasma lipids. The authors of the paper
concluded that the effect on
arterial compliance seen in this group was
about the same as would be
achieved using conventional hormone replacement
therapy.
Ref: Arterioscler Thromb Vasc
Biol 1997;17:3392-8
COMMENT
These four studies highlight
some of the excitement and also the
difficulties involved in
assessing just what phytoestrogens do to the
cardiovascular system in real
life, as distinct from the laboratory.
The first is a particularly
large and well conducted example of a number of
epidemiological studies which
suggest that soy consumption is associated
with lower lipid levels,
particularly seen in Asian cultures. The second
study is part of a great body
of data confirming that soy protein affects
lipid profiles favourably (for
a review of this subject, see J. Am. Diet.
Assoc. 1991;91:820-827).
However, one important question
is whether this effect comes from the soy
food (and if so which component
of the soy) or can be seen with other
sources of phytoestrogen, such
as the plant based isoflavones used in the
Australian lipid study. These
researchers failed to demonstrate any such
effect, despite running a well
controlled trial in normo-lipaemic adult
subjects.
Finally the second Australian
study suggests that isoflavones taken in a
pharmacological dose are
effective at reversing the decreasing elasticity
of aging female arteries around
the age of menopause.
There could be any number of
possible reasons why the first Australian
study failed to find a positive
effect of isoflavones on lipid levels.
For example:
·
The effect
seen with soy feeding is related to the protein/fat composition, not the
phytoestrogens.
·
Phytoestrogens
need to be given in larger doses than those used in this trial, or perhaps a
different isoflavone ratio would be more effective.
·
Any possible
effect was blunted by the nature of the study population, (e.g. mixing men and
women, their initially normal lipid values)
·
The effect
is subtle and needs a larger study size to uncover it.
There is certainly debate about
whether doses equivalent to those
ostensibly consumed in Asian
cultures (on average ~ 50 mg/day) are enough
to see the full pharmacological
benefits of isoflavones, or whether a
higher dose (at least 80mg or
more) is needed. Note that the Australian
lipid study used 55 mg/d,
whereas the arterial compliance study gave 80mg/d.
There is also a view that the
effects of soy are mainly seen in
hyperlipidaemic subjects (e.g.
see the article in N. Engl. J. Med. 1995;
333:276-282).
Whatever the case, there is
much interest in store in watching our
knowledge unfold on the impact
of soy and phytoestrogens on cardiovascular
health.
trials.
Return to index
ISSUE #
49: Homocysteine and the heart
Study one
The true impact of homocysteine on
cardiovascular disease is suggested by
the latest findings to be released from the
Nurses Health Study. This shows
that higher intakes of folate and vitamin
B6 are associated with
significantly lower incidence of
cardiovascular disease (both fatal
coronary heart disease and non-fatal
myocardial infarction).
This effect showed a dose-effect
relationship, and was seen whether the
source of the vitamin is food or
supplements.
More than 80,00 women were enrolled in the
Nurses Study in 1980, at which
time they completed a food frequency
questionnaire and were then followed
up for 14 years.
The table below shows the extent of the
reductions in CHD seen at various
levels of nutrient intake, after
controlling for a wide variety of other
possible influences, the strongest of which
were smoking and intake of
vitamin E.
The relationship between CHD and folate was
greatest in women who consumed
alcohol at least once per day. Adding B12,
riboflavin or methionine intake
into the models did not significantly alter
the outcome.
Overall, women who used multiple vitamin
supplements had a 25% reduction in
risk of CHD ((RR=0.76, 95% CI 0.65-0.90).
However, although supplements
were the greatest contributors to intake of
both folate and B6 in these
subject's diets, the relationships between
nutrient intake and CHD were
still present to much the same degree in
those who did not use supplements.
Table : Relative risk of cardiovascular
disease for various quintiles of
average daily
nutrient intake
Quintile: 1 2 3 4 5
----------------------------------------------------------------------------------------------
FOLATE p=0.003
Intake: 158 217 276 393 696 µg
RR: 1.0 0.86 0.86 0.78 0.69
(95%
CI) (.7-1.05) (.7-1.06)
(.63-.98) (.55-.87)
VIT.B6 p=0.002
Intake: 1.1 1.3 1.7 2.7 4.6 mg
RR: 1.0 0.92 0.86 0.88 0.67
(95%
CI) (.76-1.12) (.7-1.05) (.76-1.10) (.53-.85)
Ref: JAMA. 1998;279:359-364
Study two
Plasma homocysteine is significantly
elevated in teenage women with
anorexia nervosa,, and lowers during
nutritional treatment, according to
the results of a recent Spanish study.
Forty three female adolescent patients
attending the tertiary children's
Hospital Sant Joan de Déu had their homocysteine
levels measured.
A third of the patients were found to have
homocysteine values above what
was considered the reference range, whilst
a half had levels described as
`high normal'. Three quarters of the abnormally low values lowered after
nutritional treatment. With only one
exception, serum vitamin B12 and
folate levels were normal in all patients.
Ref: Eur J Clin Nutr 1998;52; 172-5
Study three
Supplementing breakfast cereals with folate
decreases plasma homocysteine
levels, but the dose required for optimal results may be higher than
currently recommended, according to an
American study.
Varying levels of fortification resulted in
decreases in homocysteine (and
corresponding increases in plasma folate as
follows:
|
Folate content
|
Decrease in
plasma homocysteine
|
Significance
|
|
127 µg
|
3.7% (p = 0.24)
|
p = 0.24
|
|
499 µg
|
11.0 % (p<0.001)
|
p<0.001
|
|
665 µg
|
14.0% (p = 0.001)
|
p = 0.001
|
Ref:
NEJM 1998; 338: 1009-1015
Study four
Fortification of breakfast cereal with 200
µg of folic acid per serve
significantly decreased plasma homocysteine
concentration in a general
practice population with an initially low
intake of folic acid.
The result of this level of fortification
was a significant increase in
both serum and red cell folate and a 10%
decrease in plasma homocysteine (
p <0.001). This was mainly seen in those
who initially had the highest
plasma homocysteine or the lowest serum
folate.
Ref: Eur J Clin Nutr 1998,52: 407 - 411
COMMENT
These four studies taken collectively
provide a significant advance in our
understanding of the homocysteine story.
The results from the Nurses study reinforce
the impression - already taking
root when we last reviewed this subject -
that in Western populations, it
is the folate levels which are the most
significant in terms of
homocysteine and cardiovascular disease.
Vitamin B6 also proved to be
significant in this study, but vitamin B12
levels were not.
This study is important in being the first
to give a comprehensive
epidemiological analysis in a large
population of all the main nutrients
involved in the homocysteine pathway.
However, it is also a study of
healthy Western adult women. Whether the
same result would apply to groups
with more marginal B12 status (e.g. the
elderly and vegetarian populations)
remains to be seen.
It is also important in establishing that
dietary sources of folate and B6
may be just as useful as supplements in
offering protective effects. This
stands in direct contrast to the findings
of the Nurses study in relation
to vitamin C (cataracts) and vitamin E
(heart disease).
The second study confirms that in one high
risk group for malnutrition -
anorectic teenage girls - homocysteine levels can be elevated, even
before
there is any detectable drop in folate or
B12 status.
This holds out the intriguing possibility
that plasma homocysteine readings
might provide a valuable early stage marker
of some elements of the process
by which poor or sub-optimal nutrition
leads to cardiovascular disease.
The final two studies refine our
understanding of what sort of increases in
folate intake might be needed to reduce
homocysteine levels in clinical
practice. It appears that, whilst moderate
doses may be effective in a
population whose prior consumption is low,
in a general population higher
doses are necessary.
Admittedly homocysteine issues may not be
the primary rationale for folate
supplementation around the world at present
- it is much more the
prevention of neural tube defects. But this
is still something we have to
work out on the path to conducting the
necessary intervention clinical
trials.
Return to index
