Bridging the Gap: Conference Record [Abstract book, International Conference on AIDS (12th: 1998: Geneva, Switzerland)]

840 Abstracts 42254-42334

12th World AIDS Conference

Methods: Subject were recruited from the hospital's HIV clinic into a randomized double-blinded eight-week study and given either a placebo or an amino acid
mixture containing 7 g arginine, 7 g glutamine, and 3 g HMB daily (AA/HMB). Lean
body mass and fat mass were measured by an air displacement plethysmograph
upon enrollment and at 4, and 8 weeks.
Results: Sixteen subjects have completed the 8-week protocol. (AA/HMB mixture, n = 8, Placebo, n = 8). At 8-weeks the subjects consuming AA/HMB mixture
had gained 3.74 ~ 1.37 kg while the placebos had gained 1.22 ~ 1.36 kg. In the
AA/HMB most of the gain was lean body mass (2.13 ~ 1.62 kg) while the placebos
did not gain any lean (-0.26 ~ 1.63). In turn, the AA/HMB mixture gained 1.60
~ 1.70 kg of fat and the placebos gained 1.49 ~ 1.70 kg. The AA/HMB mixture
increased absolute CD4 numbers by 31.8 ~ 45.4 cells/mm3 vs 1.2 ~ 39.3 and %
lymphocytes by 6.14 ~ 5.54% vs -2.38 ~ 5.19%, respectively.
Conclusion: These preliminary data indicate that the AA/HMB mixture can
markedly alter the course of lean tissue loss in patients with HIV associated
wasting. The gain in body weight with the AA/HMB mixture was primarily an
increase in lean tissue in contrast to placebos which gained fat tissue. In addition,
the immune system is positively supported by the AA/HMB mixture.
170*/42331 Vitamin A supplements and mortality among HIV
positive and negative children in Tanzania
Wafaie Fawzi1, R.L. Mbise2, F. Kawau2, M.G. Herrera1, E. Hertzmark1,
M. Fataki2, D. Spiegelman1. 1HSPH 665 Huntingto Avenue, Boston, MA, 02115,
USA; 2 Muhimbili Medical Centre, Dar-Es-Salaam, Tanzania
Objectives: To determine whether vitamin A supplements result in reduced
mortality among HIV infected and uninfected children, using a randomized, double-blind, placebo-controlled trial.
Methods: Starting in 04/1994 we randomized 687 children aged 6 mo to 5
y who were admitted to hospital with pneumonia. Children who were severely
malnourished or with clinical signs of vitamin A deficiency were excluded for
ethical reasons. At baseline, children received placebo or 400,000 IU (or half
that for infants) of vitamin A, in addition to standard treatment for pneumonia.
They received further doses of the same regimen 4 and 8 months after hospital
discharge. Follow up was for at least 1 year. Sera from children were tested for
HIV antibodies using ELISA and Western Blot tests. For positive children under
15 mo of age, HIV infection is being confirmed using a virologic test because
some (particularly 6-9 mo old) may have had maternal antibodies. Results using
the latter test will be presented at the meeting.
Results: Of 687 children, 90 deaths occurred during the study, and 72 had positive HIV tests. Compared with negative children, all-cause mortality was higher
among HIV-positive children (p < 0.001), so was mortality due to pneumonia
(p < 0.03), diarrhea (p < 0.001), and HIV/AIDS (p < 0.001). Among positive
children, vitamin A supplements reduced all-cause mortality by 49% (RR = 0.51,
p < 0.04), but had no effect on all-cause mortality among negative children (RR
= 0.94, p = 0.81). Among positive children, 21% of those on placebo experienced
a diarrhea-related death, compared to none of those who received vitamin A
(Fisher's exact p < 0.003). An 86% reduction in diarrhea-related mortality (p <
0.03) among negative children was observed. Supplements also resulted in a
significant reduction in the severity of diarrhea among surviving children. Results
on this and other morbidity endpoints will be presented. Potential mechanisms of
action of vitamin A in HIV infection and policy implications of these findings will
be discussed.
Conclusion: Vitamin A deficiency is common among children in many developing countries, and is particularly severe among HIV-positive children and adults.
In the last year, 460,000 children died of AIDS worldwide, most of whom are in the
developing world. Our findings indicate that vitamin A supplements, a low-cost
intervention, reduce mortality of HIV-infected children.
42332 1 Whole body protein turnover in HIV-infected patients,
effect of nutritional intervention
Kaspar Berneis1, M. Battegay2, S. Bassetti2, A. Leisibach2, R. Nuesch2,
A. Tietz2, U. Keller1. 'Dep. of Research University Hospital, Basel; 2University
Hospital of Basel, Basel, Switzerland
Objective: Weight loss and protein malnutrition represent major complications in
HIV-infected patients. We evaluated the effect of an oral nutritional supplement
combined with nutritional counselling on whole body protein turnover.
Methods: HIV-infected individuals (n = 15) with CD4+ cells <500 or body mass
index <21 kg/m2 in stable clinical condition and without change of antiretroviral
therapy were randomly allocated into one of two groups during 12 weeks. One
group (n = 7) received oral nutritional supplements and dietary counselling by a
dietician. The supplement contained 600 kcal, complete macro- and micronutrients. The other group (n = 8) was monitored identically but no supplement and
no specific nutritional advice was given (control group). Lean mass, fat mass
and leucine oxidation rate, a parameter of whole body protein catabolism, were
determined.
Results: Lean mass, determined by bioelectrical impedance analysis, as per
centage of total body weight increased in the oral supplement group from 83.5
~ 1.8 to 86.2 ~ 1.7% (p < 0.05) but remained unchanged in the control group
(from 82.9 ~ 1.7 to 84.5 ~ 2.5%). Fat mass as percentage of total body weight
decreased in the oral supplement group from 16.5 ~ 1.78 to 13.8 ~ 1.7% (p <
0.05) but remained unchanged in the control group (from 17.1 ~ 1.6 to 15.5 ~
2.5%). The Leucine oxidation rate measured by 1-13C-leucine infusion technique)
declined in the group which received oral supplements (from 0.327 ~ 0.023 to

0.2643 ~ 0.018 / mol/kg/min) (p < 0.05 vs control group) but remained unchanged
in the control group (from 0.311 ~ 0.0274 to 0.339 ~ 0.021 lpmol/kg/min). Endogenous leucine appearance, a parameter of whole body protein breakdown,
remained unchanged in both groups (intervention group from 1.92 ~ 0.19 to 1.73
~ 0.135 u/mol/kg/min; control group from 2.21 ~ 0.16 to 2.27 ~ 0.14).
Conclusions: These findings demonstrate an acute anabolic response in HIV
infected subjects due to nutritional supplements and counselling. Nutritional therapy is important to decrease whole body protein catabolism and to increase lean
body mass in HIV-infected patients.
S42333 Body cell mass (BCM) in HIV-infected (HIV+) males in the
community programs for clinical research on AIDS
(CPCRA) in 1996-1997
Norma Muurahainen1, G. Collins2, D. Wheeler3, N. Mateo4, M. Madans5,
G. Bartsch2, C. Gilbert3. 1Serono Laboratories, Norwell, MA 02061; 2CPCRA,
Minneapolis, MN; 3CPCRA, Washington, DC; 4CPCRA, Detroit, Ml; 5CPCRA,
Bethesda, MD, USA
Objectives: To determine whether a cross-section of HIV+ males with CD4+
< 200 cells/mm3 and documented weight-stable have lower BCM compared to
males from the Third National Health and Nutrition Examination Survey (NHANES
III), and whether these differences vary by percent desirable weight.
Methods: Measurements of body weight and composition were obtained from
455 HIV+ males aged 20-77 years enrolled in a CPCRA nutrition study (8/96
to 11/97) and 4310 males from NHANES III aged 20-70 years (controls). Both
groups had BMI 15-29 kg/m2 and body composition determined by single frequency bioelectrical impedance analysis. Desirable weights were derived from
NHANES I. Multiple regression on the logarithmic scale was used to adjust for
covariates.
Results: Race/ethnicity (HIV+, controls): African American (37%, 29%);
Latino/Hispanic (15%, 34%); white (48%, 36%); mean age-years (41, 41); mean
weight-kg (72.3, 74.4), mean BCM-kg (28.2, 31.2); mean percent desirable weight
(93.9, 98.7). HIV+: 55% with history of AIDS-related illnesses, CD4+ count ranging 0-199 (mean 103) cells/mm3; 96% were on antiretrovirals, including 74% on
a protease inhibitor.

Mean BCM (kg) by percent des
Percent desirable weight
HIV+
Controls
Adjusted percent differencet;irable weight
<90% 90%-95%
N Mean N Mean
163 25.8 96 28.0
993 28.2 543 30.3
90.2 90.8

95%N
114
1358

105%
Mean
29.8
31.5
91.9

-105%
N Mean
82 31.1
1416 33.3
92.0

tAdjusted for race, age, age2, height, and body weight. 'Significantly different at p =.0001
Conclusions: These weight-stable HIV+ men with CD4 < 200 had lower
BCM than the controls at all levels of desirable body weight. BCM differences
may reflect metabolic abnormalities that persist despite combination antiretroviral
therapy.
42334 | Impact of HIV and substance use on infant birthweight
and weight gain during pregnancy
Maike Rahn1, Wenho Yang2, Shaikh Ladan4, Gipeda Peter3, Wafaa EI-Sadr3,
Elaine Abrams3, Subhasree Sai Raghavan3. 1St. Lukes-Roosevelt Hospital
Center, 573, 10th Street, Brooklyn, NY; 2Columbia University; 3Harlem Hospital,
New York, NY; 4 Temple University Medical School Philadelphia, PA, USA
Objectives: To determine the association between HIV status, substance use
and weight (Wt) gain during pregnancy and its impact on infant birth weight.
Methods: A retrospective analysis of weight gain during pregnancy was conducted on 21 HIV-infected (HIV P) and 21 HIV uninfected (HIV N) pregnant
women receiving primary health care at Special Prenatal and Infectious Diseases
Clinics of Harlem Hospital Center in New York City. Weight gain was calculated
from pre pregnancy weight and weight at the last visit during pregnancy. Net Wt.
Gain was calculated as Wt Gain in Ibs - (Birth wt in gm/1000) 2.2.
Results: Characteristics: African American/African (70%) and Hispanic/Latino
(30%), mean age in years (HIV N, HIV P) (28, 29), substance use (0%, 76%),
mean CD4+ of HIV P women (426 cells/mm3).
HIV Status GA (wks) PPBMI (kg/ht2) Wt Gain (Ibs) Net Wt Gain(lbs) Birth Wt (gm)
(n) (n) (n) (n) (n)*
HIV N 38.9 ~ 1.8 24.6 + 4.7 28.2 ~ 11.1 23.8 ~ 10.2 3230 ~ 519
(14) (17) (17) (10) (12)
HIV P 38.0 ~2.6 26.2 1 8.1 26.8 + 9.1 20.2 ~ 9.3 2884 ~ 547
(17) (18) (18) (17) (20)
p = 0.05, NS = Not Significant
Conclusion: HIV P women had slightly lower weight gain during pregnancy
than HIV N women (NS). Infants born to HIV P women had significantly lower
birth weights than HIV N women (p =.054). However, these lower birth weights
appear to be not mediated through lower weight gain during pregnancy. We
speculate that high incidence of substance use (70%) among our HIV P women
may have mediated the low birth weights. We were unable to evaluate the impact
of substance use on weight gain and birth weight due to lack of adequate non
substance using HIV P and substance using HIV N women.