European Journal of Clinical Nutrition (1999) 53, 165±173
ß 1999 Stockton Press. All rights reserved 0954±3007/99 $12.00
http://www.stockton-press.co.uk/ejcn
Nutritional status of HIV-1 seropositive patients in the Free State
Province of South Africa: Anthropometric and dietary pro®le
A Dannhauser1, AM van Staden2*, E van der Ryst3, M Nel4, N Marais1, E Erasmus1, EM Attwood 2,
HC Barnard5 and GD le Roux2
1
Departments of Human Nutrition, 2Internal Medicine, 3Virology, 4Biostatistics and 5Chemical Pathology, University of the Orange Free
State, Bloemfontein, South Africa
Objective: To evaluate the nutritional intake and status of HIV-1 seropositive patients, as well as the relationship
between malnutrition and disease stage.
Design: A cross-sectional study.
Settings: The Immunology Clinic at the Pelonomi Hospital in Bloemfontein, South Africa.
Subjects: Eighty-one HIV=AIDS patients in different stages of disease were recruited consecutively from
January to May 1995. Eleven of these patients were followed in 1997.
Main outcome measures: Anthropometric data including current weight, height, triceps skinfold thickness, midupper-arm circumference, body mass index and bone-free arm muscle area were collected. Nutrient intake was
estimated using a diet history in combination with a standardised food frequency questionnaire. The patients
were divided into 3 groups according to their CD4‡ T cell counts.
Results: The men were leaner (BMI ˆ 18.9) than the women (BMI ˆ 22.7) and patients with a CD4‡ T cell
count < 200 (stage III) tended to have the lowest median values for all anthropometric measurements. More than
half the patients had a low intake (< 67% of the recommended dietary allowances) of vitamin C, vitamin B6,
vitamin D, vitamin A, calcium, iron and zinc.
Conclusions: The results con®rms that HIV=AIDS patients from this population are malnourished. There was,
however, no association between disease stage and nutritional status. Nutritional supplementation of HIV=AIDS
patients should be considered, as this might lead to improved immune function in these patients.
Descriptors: anthropometry; dietary intake; HIV infection; South Africa
Introduction
More than 30 million people worldwide are currently
infected with HIV-1. Of these, more than 90% live in the
developing world, and 20.8 million reside in sub-Saharan
Africa (UNAIDS, 1998). The HIV-1 epidemic in South
Africa started in the white homosexual population in the
mid-1980s, but by 1990 it was evident that heterosexual
contact was becoming the predominant mode of transmission in the country, with a subsequent explosive epidemic
in the heterosexual population (ADSA, 1997). It is estimated that more than 1000 new infections occur daily in
South Africa, and that more than 3 million South Africans
were already infected with HIV-1 by the end of 1997
(Department of Health, South Africa, 1998). The seroprevalance of HIV-1 antibodies in pregnant women attending public antenatal clinics in South Africa was 16.0% in
1997, which is a signi®cant increase from the 10.4% of
*Correspondence: Dr AM van Staden, Dept. of Internal Medicine (G73),
University of the Orange Free State, PO Box 339, Bloemfontein 9300,
South Africa.
Received 21 April 1998; revised 14 September 1998; accepted 21
September 1998.
1995 and 14.1% of 1996 (Department of Health, South
Africa, 1998). In the Free State the seroprevalence in
antenatal clinic attendees at the end of 1997 was 19.6%.
Nutritional de®ciencies are well-documented complications of HIV infection (Dworkin et al, 1985; Cuff, 1990;
Hecker & Kotler, 1990). HIV=AIDS patients develop
severe protein and energy malnutrition as demonstrated
by a depletion of body fat and muscle mass. De®ciencies of
several vitamins and minerals also develop. The pathogenesis of malnutrition in AIDS is multifactorial and includes
decreased food intake, decreased nutrient absorption and
decreased ef®ciency of utilisation, combined with increased
nutritional needs and increased tissue metabolism. An
association between physiological micronutrient de®ciences and disease progression was demonstrated in westernised populations (Baum et al, 1995), while a possible
relationship between dietary intake and disease progression
was also shown (Dworkin et al, 1990). Little is known,
however, regarding the association between dietary intakes
and disease progression in HIV=AIDS patients from developing populations, and further studies, especially in the
African context, are recommended (Castetbon et al, 1997).
The areas hardest hit by the AIDS epidemic are subSaharan Africa, south Asia and south-east Asia, where
the prevalence of micronutrient de®ciencies is generally
higher (Semba, 1997). The all-cause mortality was also
Nutritional status of HIV-1 positive patients
A Dannhauser et al
166
much higher and the progression time to death in AIDS
patients shorter in rural Uganda than in developed countries
(Morgan et al, 1997).
A further aspect that should be kept in mind is that
HIV=Mycobacterium tuberculosis co-infection often occurs
(Niyongabo et al, 1994; Leroy et al, 1997), further increasing the nutritional burden of the patients (Robinson et al,
1986). HIV=AIDS patients also have a signi®cantly higher
risk of contracting TB infection owing to the underlying
immune de®cincey (Wilkinson & Moore, 1996). Mycobacterium tuberculosis (TB) infection is a major public health
problem in South Africa, especially among the lower
socioeconomic segment of society (Taylor & Benatar,
1989) from which a large number of HIV=AIDS patients
also come, and this has resulted in co-infection with HIV
becoming another emerging epidemic in South Africa
(Wilkinson & Davies, 1997).
The staple diet of the majority of people in the Free
State province is maize, and a large number of them are
from lower socioeconomic groups. Although maize meal
and bread are enriched with certain vitamins and minerals,
the adequacy of this in patients with increased nutritional
needs is unsure. A signi®cant number of HIV=AIDS
patients in the Free State are, therefore, possibly at risk
for malnutrition, which may contribute to disease progression in this group. Previous studies of samples from the
adult Free State population, however, did not reveal
extreme nutritional de®ciencies. The men tended to be of
normal weight (mean BMI for different age groups 22.3±
24.7), while the women were mainly overweight (mean
BMI 27.1± 31.1) (Mollentze et al, 1995). Their diets were
generally adequate, with few micronutrients having a mean
lower than the recommended dietary allowances (RDA) of
the United States (N. Silvis & W. F. Mollentz, 1995,
unpublished). Pregnant women from the population had
weights mainly in the normal to high pregnancy ranges,
while their diets were inadequate in various micronutrients
(Dannhauser et al, 1996a). On the other hand, the nutritional status and dietary intakes of pre-school-age black
children in the Bloemfontein district (Dannhauser et al,
1996b) and of informal settlement areas near Bloemfontein
(Dannhauser et al, 1998) indicated a moderate prevalence
of underweight and stunting, and inadequate intake of
energy and various micronutrients, while the protein was
mainly of a low quality.
The aim of the present study was to evaluate the
nutritional intake and nutritional status and its association
with disease stage, of a group of HIV-1 seropositive
patients from the Free State.
Patients and methods
A cross-sectional descriptive study was carried out. Eightyone consecutive HIV-1 seropositive patients attending the
Immunology Clinic at the Pelonomi Hospital, Bloemfontein, South Africa, were recruited from January 1995 to
May 1995. The patients attending this clinic are from a low
socioeconomic environment. In 1997 a follow-up study was
conducted. For ®nancial reasons, it was decided to follow
only 45 of the original 90 patients. We therefore selected a
group of 35 who had both dietary and laboratory evaluations in 1995, as well as 10 patients who were followed up
in 1996. Of the 45 patients, we were able to trace only 16
patients in 1997. Of the rest, 9 had died, 9 had moved to
another town, for 2 we had no home address, and the other
9 were untraceable. Of the 16 that did attend the clinic for
follow-up, only 11 were evaluated with regard to diet and
anthropometric values. Approval from the Ethics Committee of the Health Sciences Faculty of the University of the
Orange Free State was obtained, and all patients gave
informed consent for inclusion in the study. Demographic
details (age, gender and residential area) were obtained and
patients were divided into three groups according to their
CD4‡ T cell count, as determined by ¯ow cytometry (Van
der Ryst et al, 1998); group I ( 500 cells=mm3), group II
(200 ±499 cells=mm3) and group III (0 ±199 cells=mm3).
Dietary and anthropometric evaluation
Dietary and anthropometric evaluations were completed for
81 patients. Usual dietary intake was estimated by trained
dietitians using the 24 h recall method, in combination with
a standardised food frequency questionnaire covering the
month before the interview. If the previous 24 h were not
representative of a usual day, the respondent had to indicate
the variation. The results from a food frequency questionnaire have been shown to be similar to those from a dietary
history (Feunekes et al, 1993), therefore the food frequency
information was used in combination with the 24 h recall to
estimate the mean intake on a usual day for each respondent. Interpreters were used for patients who did not speak
English or Afrikaans. Aids such as food models were used
to decrease the accuracy of the size of food portions
described by the patients. The patients were asked to
demonstrate how much of a speci®c food stuff they
normally ate. The portion sizes were estimated in household measures and converted to grams using the conversion
®gures in the NRIND Food Quantities Manual (Langenhoven et al, 1986). The dietary data were analysed by
computer, using the NRIND Food Composition Tables
(Langenhoven et al, 1991). Nutrient intakes were compared
to the recommended dietary allowances (RDA) of the
United States (Food and Nutrition Board, 1989). A value
of less than 67% of the RDA was considered to be
inadequate. It must, however, be noted that nutrient
requirements of HIV=AIDS patients are higher owing to
the catabolic nature of the disease. Therefore, as the exact
nutritional requirements for these patients are not known,
the RDA was used as a standard of reference. The RDAs
were chosen instead of those compiled by the World Health
Organisation (WHO) (Ralph, 1993) in order to enable
comparison of the results of this study with those of other
studies.
Anthropometric measurements included body weight,
height, triceps skinfold thickness (TSF) and mid-upperarm circumference (MUAC). Bioelectrical impedance analysis (BIA) to determine body fat and fat-free mass (FFM)
was not used in this study because the technology was not
available on site. TSF and MUAC were measured, as they
were considered to be the least invasive method for the
patients at this busy outpatients clinic, and upper-arm bonefree muscle area (UBMA) was used to estimate muscle
reserves or lean body mass (LBM). Furthermore, although
change in FFM was not measured in the present study, it is
interesting to note that skinfold thickness (SF) appears to be
superior to BIA for detecting fat-free mass (FFM) change
in patients with AIDS (Paton et al, 1996).
All anthropometric measurements were done by a
trained dietitian, according to standard methods (Frisancho,
1990; Lee & Niemann, 1993). Body weight was measured
using a calibrated Seca scale, weighing up to 150 kg and
Nutritional status of HIV-1 positive patients
A Dannhauser et al
accurate to 0.1 kg. Patients were weighed barefoot and in
light indoor clothing. Their height was measured to the
nearest centimetre by means of an anthropometer, measuring up to 2 m and with a metal sliding headpiece attached to
the scale. Patients were measured barefoot and stood on the
metal platform with their heels touching the anthropometer.
MUAC was measured with a steel measuring tape accurate
to the nearest 0.1 cm. TSF was measured using a Harpenden skinfold caliper with a standardised pressure of
10 g=mm2, measuring to the nearest 0.2 mm. Upper-arm
bone-free muscle area (UBMA) was calculated using the
formula (Frisancho 1990):
UBMA …cm2 † ˆ f‰MUAC …cm† ÿ piTSF …cm†Š2 g=
4p ÿ 10 cm2
for males
2
UBMA …cm† ˆ f‰MUAC …cm† ÿ pTSF …cm†Š2 g=
4p ÿ 6:5 cm2
for females
The values of the National Centre for Health Statistics
(NCHS) were used as reference values (Frisancho, 1990).
Values below the ®fth centile of the NCHS for TSF,
MUAC and UBMA were considered to be depleted (Frisancho, 1990). Body mass index (BMI), or Quatelet index,
was calculated using the formula (Frisancho, 1990):
(CI) were calculated for the mean, median or percentage
differences between the three groups of patients with
respect to nutrient intakes and anthropometric values.
Results
Nineteen of the 81 patients (23.5%) were male, and
76.5% were from urban areas in the Free State. The
high percentage of women represents the demographic distribution of HIV-infection prevalence in SA.
The median age of the patients was 34 years (range 16 ±69
years). There were 23 patients in group I, 35 in group II,
and 23 in group III. The clinical diagnoses of the patients at
the time of dietary evaluation comprised the following:
tuberculosis 20 (24.6%), herpes zoster 5 (6.2%), gastroenteritis 4 (5%), syphilis 6 (7.4%) and lymphadenopathy 6
(7.4%). A further 23 (28.3%) patients were completely
asymptomatic, while two patients (2.5%) had more than
one clinical problem.
BMI values of less than 20 for men, and 19 for women
were considered to be depleted (Bray, 1978).
Weight loss was not calculated as the patients could not
recall their pre-illness weights.
Anthropometric data
The men (median BMI 18.9) were leaner than the
women (median BMI 22.7) (Table 1). Patients in group
III (CD4‡ T cell count < 200 cells=mm3) tended to have
the lowest median values for all the anthropometric variables (Table 1). Furthermore, more patients in group III had
depleted ( < 5th centile) values for MUAC and UBMA
(Table 2). The 21 patients who were infected with Mycobacterium tuberculosis tended to have lower mean anthropometric values, but these values were not statistically
signi®cant.
Statistical analysis
All statistical analyses were performed using SAS statistical software (SAS, 1990). Frequencies and percentages
were used to describe categorical data, and means, standard
deviations, medians, 25th centiles and 75th centiles for
continuous data. Ninety-®ve percent con®dence intervals
Dietary data
The median protein (60.9 g=d) and energy (8719 kJ=d)
intakes of most patients were within normal ranges, with
only 12.3% showing inadequate (< 67% of the RDA)
protein and 24.7% inadequate energy intake (Table 3).
More than 50% of the patients had inadequate intakes of
vitamin C (75.3%), vitamin B6 (64.2%), vitamin D
BMI …kg=m2 † ˆ Weight …kg†=‰height …m†Š2
Table 1
Mean, standard deviation (s.d.), 25th centile, median and 75th centile of anthropometric criteria of HIV-1 seropositive patients
Anthropometric criteria a
BMI (body mass index)
Males
Females
Group I
Group II
Group III
Total
MUAC (mid-upper-arm circumference)
Group I
Group II
Group III
Total
TSF (triceps skinfold)
Group I
Group II
Group III
Total
UBMA (upper bone-free muscle area
(Group I)
(Group II)
(Group III)
Total
a
N
Mean
s.d.
25%
Median
75%
19
61
22
35
23
80
20.21
23.06
22.7
23.0
21.1
22.4
3.37
4.89
4.4
4.5
5.2
4.7
17.86
19.68
19.0
20.1
17.1
19.0
18.93
22.68
21.8
22.7
19.7
21.4
22.82
36.17
25.8
26.2
23.9
25.7
23
35
23
81
26.7
26.5
25.0
26.2
4.1
6.1
5.6
5.4
23.8
24.6
20.8
23.6
26.2
27.0
24.3
26.2
29.2
29.7
28.3
29.2
22
35
23
80
22
35
23
80
1.65
1.72
1.5
1.63
30.6
31.7
26.5
29.9
0.9
0.81
0.87
0.85
9.2
13.6
15.4
13.1
0.88
1.1
1.07
1.1
23.5
20.9
15.1
19.8
Group I, CD4‡ 500 cells=mm3; group II, CD4‡4 200 ± 499 cells=mm3; group III, CD4‡ < 200 cells=mm3.
1.6
1.67
1.48
1.56
31.2
31.2
21.8
30.1
2.28
2.05
1.75
2.05
37.7
36.2
36.6
36.6
167
Nutritional status of HIV-1 positive patients
A Dannhauser et al
168
Table 2 Body composition depletion according to the stage of disease
(a) Differences in proportions of patients with anthropometric values below the 5th centile of the National Centre for Health Statistics (NCHS) in the
different patient groups, as calculated by 95% con®dence intervals (CI)
Difference in proportion < 5th centile
< 5th centiles
Anthropometric criteria
N
MUAC (Mid upper arm circumference)
Group I
23
Group II
35
Group III
23
Total
81
TSF (Triceps skinfold)
Group I
23
Group II
34
Group III
23
Total
80
UBMA (Upper bone-free muscle area)
Group I
23
Group II
34
Group III
23
Total
80
n
%
5
11
9
25
21.7
31.4
39.1
30.9
23
33
23
79
100
97.1
100
98.8
4
12
11
27
17.4
35.3
47.8
33.8
(95% CI)
Group III ± II
(95% CI)
Group III ± I
(95% CI)
Group II ± I
[717.5; 32.9]
[78.7; 43.5]
[713.1; 32.5]
[72.7; 8.6]
[77.9; 7.9]
[78.6; 2.7]
[713.4; 38.5]
[4.8; 56.1]*
[74.4; 40.2]
(b) Differences in proportions of patients with values below the lowest normal BMI value in the different groups, as calculated by 95% CI
< 19 : F; < 20 : M
BMI (body mass index)
n
n>
%
Group I
Group II
Group III
Total
22
35
23
80
6
7
9
22
27.3
20.0
39.1
27.5
Difference in proportions with low BMI
Group III ± II
Group III ± I
Group II ± I
[74.8; 43.1]
[715.4; 39.1]
[730.1; 15.6]
Group I, CD4‡ 500 cells=mm3; group II, CD4‡ 200 ± 499 cells=mm3; group III, CD4‡ < 200 cells=mm3.
F ˆ female; M ˆ male.
* Signi®cant difference P < 0:05.
(64.2%), calcium (60.5%), iron (59.3%), vitamin A (58%)
and zinc (50.6%). Forty-eight per cent of the patients,
furthermore, had an inadequate niacin intake (Table 3).
Patients in group III tended to have the lowest nutrient
intakes of the three groups (Table 3). The nutrient intakes
that differed signi®cantly between the groups according to
the 95% CI are summarised in Table 4. According to the
95% CI, the median intakes of calcium, potassium, ribo¯avin and vitamin A of group III patients (CD4‡ T cell
count < 200 cells=mm3) were signi®cantly lower than for
group I (CD4‡ T cell count 500 cells=mm3) (Table 4).
Some of the 95% CI were extremely wide, however,
making a ®rm conclusion dif®cult. Data on the laboratory
analysis of 35 of these patients and the correlation with the
nutrient intake are reported in a separate paper (Van Staden
et al, in press).
Twenty-one of the 81 patients suffered from M. tuberculosis infection at the time of evaluation, and their nutrient
intake tended to be lower than that of the uninfected
patients, but was not statistically signi®cant, except for
vitamin A. The dietary intake of the group III patients
infected with M. tuberculosis, however, tended to be higher
than the uninfected group III patients.
In order to evaluate a possible deterioration in nutritional
status of patients with disease progression, it was attempted
to follow the patients after 2 years. Unfortunately, data from
only 11 patients were available in 1997. They showed a
tendency towards higher intakes of most nutrients, while
their anthropometric values tended to be the same or
slightly lower. According to the 95% CI, the changes
were not signi®cant, but this may be due to the small
number.
Discussion
Malnutrition is an important consequence of infection with
HIV-1. Weight loss and decreases in food intake are
prominent features of HIV-associated disease (Kotler et
al, 1989; Summerbell, 1994; Eldridge, 1996). Decreases
in body weight, body fat percentage and BMI are viewed as
the earliest indications of deterioration of nutritional status
in HIV=AIDS patients (Mcorkindale et al, 1990). In fact,
AIDS was once commonly known in Africa as `slim'
disease (Serwadda et al, 1985). Various mechanisms for
this deterioration in nutritional status have been identi®ed
and can be divided into four categories: decreased oral
intake, decreased nutrient absorption, increased nutrient
requirements, and changes in lipid metabolism and transport
leading to lean body wasting (Fields-Gardner, 1995;
Eldridge, 1996). De®ciencies of protein and energy intake,
as well as zinc, selenium, iron, vitamin B6, folic acid,
vitamin B12, thiamin, vitamins A, C and E (Reaidi &
Cosette, 1992; ADSA 1996), all the carotenoids, cholesterol
(Lacey et al, 1996), vitamin D (Lochner & Schneider, 1994)
and magnesium (Skurnick et al, 1996) have been reported in
HIV=AIDS patients. The present study has identi®ed a low
micronutrient intake by HIV=AIDS patients from the Free
State Region of South Africa in all three groups tested, with
a tendency towards a lower intake for group III patients.
Nutritional status of HIV-1 positive patients
A Dannhauser et al
Table 3 Mean, standard deviation (s.d.), median and percentage of HIV-1 seropositive patients in the subgroups and in total group with nutrient intakes
lower than 67% of the RDAa
Nutrient
Total energy (kJ)
Mean
s.d.
Median
% < 67%RDA
Total protein (g)
Mean
s.d.
Median
% < 67%RDA
Calcium (mg)
Mean
s.d.
Median
% < 67%RDA
Magnesium (mg)
Mean
s.d.
Median
% < 67%RDA
Phosphorus (mg)
Mean
s.d.
Median
% < 67%RDA
Iron (mg)
Mean
s.d.
Median
% < 67%RDA
Zinc (mg)
Mean
s.d.
Median
% < 67%RDA
Copper (mg)
Mean
s.d.
Median
Pottasium (mg)
Mean
s.d.
Median
Vitamin A (mg RE)
Mean
s.d.
Median
% < 67%RDA
Vitamin E (mg a-TE)
Mean
s.d.
Median
% < 67%RDA
Vitamin D (mg)
Mean
s.d.
Median
% < 67%RDA
Thiamin (mg)
Mean
s.d.
Median
% < 67%RDA
Ribo¯avin (mg)
Mean
s.d.
Median
% < 67%RDA
Niacin (mg NE)
Mean
s.d.
Group I b
n ˆ 23
Group II b
n ˆ 35
Group III b
n ˆ 23
Total
n ˆ 81
9612
4353
8719
17.4%
9316
3363
9297
20.1%
7985
3259
7962
39.1%
9022
3659
8719
24.7%
70.1
33.4
59.1
8.7%
68.4
24.1
66.3
8.6%
61.9
35.4
58.8
21.7%
67.0
30.1
60.9
12.3%
712.8
656.1
472.1
56.2%
530.9
239.4
537.6
51.4%
396.9
252.2
397.4
78.3%
544.5
418.0
450.8
60.5%
402.9
166.0
346.4
4.3%
365.0
173.3
348.4
11.4%
308.2
126.9
300.1
26.1%
359.6
161.5
335.3
13.6%
1314.5
693.4
1084.2
0.0%
1157.5
430.8
1136.6
8.6%
969.1
433.4
962.3
26.1%
1148.6
529.1
1084.2
11.1%
10.2
4.9
9.4
60.9%
9.0
4.9
8.5
65.7%
8.6
4.5
7.5
47.8%
9.23
4.76
8.22
59.3%
9.8
4.9
8.2
47.8%
9.2
3.9
8.9
51.4%
9.08
5.70
7.58
52.2%
9.3
4.7
8.4
50.6%
1.3
0.6
1.1
1.2
0.7
1.0
1.08
0.60
0.86
1.19
0.65
1.02
2992.6
1746.2
2803.6
2511.0
939.5
2396.5
2111.8
863.1
1950.0
2534.4
1237.1
2395.9
866.2
684.8
555.0
43.5%
696.9
665.5
476.0
54.3%
436.5
368.8
326.7
78.3%
671.0
617.8
436.0
58.0%
19.1
14.1
14.8
8.7%
18.9
11.3
16.2
11.4%
15.39
9.01
15.08
17.4%
17.93
11.54
15.84
12.3%
4.8
4.3
3.0
73.9%
4.0
3.5
3.5
54.3%
3.70
3.71
2.32
69.6%
4.14
3.75
3.16
64.2%
1.37
0.58
1.30
13.0%
1.29
0.66
1.26
20.0%
1.05
0.48
1.06
39.1%
1.24
0.60
1.22
23.5%
1.43
1.05
1.23
26.1%
1.15
0.50
1.03
34.3%
0.92
0.50
0.87
52.2%
1.16
0.72
1.03
37.0%
11.49
5.77
11.76
5.66
11.68
8.73
11.66
6.61
Continued
169
Nutritional status of HIV-1 positive patients
A Dannhauser et al
170
Table 3 Continued
Nutrient
Median
% < 67%RDA
Folate (mg)
Mean
s.d.
Median
% < 67%RDA
Vitamin B6 (mg)
Mean
s.d.
Median
% < 67%RDA
Vitamin B12 (mg)
Mean
s.d.
Median
% < 67%RDA
Vitamin C (mg)
Mean
s.d.
Median
% < 67%RDA
Group I b
n ˆ 23
Group II b
n ˆ 35
Group III b
n ˆ 23
10.64
47.8%
10.77
48.6%
10.10
47.8%
10.64
48.1%
243.09
132.16
207.9
4.3%
204.51
68.93
196.88
14.3%
179.07
68.22
177.84
26.1%
208.24
93.29
196.88
14.8%
1.36
1.08
1.10
47.8%
1.07
0.57
0.94
74.3%
0.95
0.49
0.90
65.2%
1.12
0.74
0.94
64.2%
4.4
3.75
3.13
17.4%
4.11
5.03
3.29
8.6%
3.10
2.55
2.11
21.7%
3.9
4.09
3.02
14.8%
48.1
55.07
24.57
69.6%
42.92
57.29
22.54
74.3%
51.37
102.3
17.0
82.6%
46.79
71.56
22.54
75.3%
Total
n ˆ 81
a
RDA ˆ US Recommended Dietary Allowances.
Group I, CD4‡ 500 cells=mm3; group II, CD4‡ 200 ± 499 cells=mm3; group III, CD4‡ < 200 cells=mm3.
RE ˆ Retinol equivalents; a-TE ˆ alpha tocopherol equivalents; NE ˆ niacin equivalents.
b
Table 4 Nutrients for which the intake showed signi®cant differences between the three groups, as calculated by 95% con®dence intervals (CI)
Nutrients
Calcium
Potassium
Ribo¯avin
Vitamin A
Group III ± II
(95% CI)
[7277.3; 715.7]*
[7880.14; 145.30]
[70.48; 0.02]
[7342.33; 36.49]
Group III ± I
(95% CI)
[7328.9;
[71204.0;
[70.6;
[ 7 640.8;
722.2]*
741.8]*
70.05]*
747.3]*
Group II ± I
(95% CI)
[7181.07;
[7767.72;
[70.36;
[7401.93;
120.54]
304.48]
0.14]
101.74]
Group I, CD4‡ 500 cells=mm3 (n ˆ 23); group II, CD4‡ 200 ± 499 cells=mm3 (n ˆ 35); group III, CD4‡ < 200 cells=mm3 (n ˆ 23).
* Signi®cant difference, P < 0.5.
The majority of patients in the present study reported
energy and protein intakes that meet at least 67% of the
RDA. This is similar to results of previous studies demonstrating that the energy and protein intake of clinically stable
AIDS and HIV-seropositive patients meet RDA standards
(Dworkin et al, 1990). In fact, some studies demonstrated
that HIV-infected patients had higher energy intakes than
HIV seronegative controls, even though their BMI was
lower (Hogg et al, 1995). However, energy intake is related
to the stage of the infection, and rapid weight loss together
with anorexia may be a forerunner of secondary infection
(Grunfeld et al, 1992). Malabsorption and=or altered metabolism probably also play an important role in the development or persistence of cachexia in AIDS. On the other
hand, it was demonstrated that progressive wasting is not a
constant phenomenon in AIDS patients (Kotler et al, 1990)
and that the development of wasting is usually due to the
presence of speci®c diseases, such as diarrhoeal disease
(Macallan et al, 1993). Other opportunistic infections and
fever may also lead to increased energy and protein needs
(Eldridge, 1996), while HIV infection might exert a direct
effect on energy balance that varies with the degree of
immunosuppression (Sharpstone et al, 1996). In contrast,
other studies suggested that a reduced energy intake, rather
than an increased expenditure, is the primary factor in HIVrelated weight loss (Grunfeld et al, 1992; Macallen et al,
1995). Moreover, decreased appetite has been described as a
major problem in many individuals with AIDS (Resler,
1988) while `anorexia of disease' has also been reported
in HIV=AIDS (Jebb, 1997). On balance, it is most likely that
a combination of both mechanisms is involved. In this
population, patients in group III tended to have the lowest
energy intakes, which together with their increased metabolic needs can be a contributing factor in immune deterioration.
Severe weight loss of 20± 30% of usual body weight has
been reported in AIDS patients (Kotler et al, 1985; Trujillo
et al, 1992), and a trend towards a decrease in body weight
and fat mass indicators was found with disease progression
(Parisien et al, 1993). Furthermore, it was demonstrated
that the mean survival of patients who had lost more than
20% of their body weight was decreased, while a body
weight of less than 66% of ideal weight was found to be
related to time of death in AIDS patients (Kotler et al,
1989). Although weight loss was not measured in this
study, 27.5% of the patients had a low BMI ( < 19 for
females and < 20 for males). The median BMI of the men
(18.9) was lower than that of the women (22.7), indicating
that the male patients tended to be leaner than the women.
This was also seen in seronegative subjects of the same
population (Mollentze et al, 1995), and also demonstrated
in HIV-infected asymptomatic patients from Abidjan (Castetbon et al, 1997). The differences in clinical status
between men and women were, however, not evaluated in
Nutritional status of HIV-1 positive patients
A Dannhauser et al
the present study. Patients in group III tended to be leaner
than those in group I, demonstrating that also in this
population weight loss increased with progression of disease. The higher percentage of patients with depleted body
fat (as indicated by TSF values < 5th centile) and lean
muscle reserves (UBMA < 5th centile), in group III further
emphasises the fact that disease progression in this population is associated with progressive wasting. Parisien et al
(1993) also found a trend towards a decrease in body
weight as well as in the fat mass indicators as the disease
progressed. The signi®cantly higher percentage of patients
in group III than in group I that showed depleted lean body
mass (UBMA < 5th centile) is in concordance with the
study of HIV-infected outpatients in Abidjan, which also
demonstrated a decrease in lean body mass as the disease
progresses (Castetbon et al, 1997). It was also shown that a
decrease in lean body mass was related to the decrease in
body cell mass (Niyongabo et al, 1997) and that body cell
mass depletion was out of proportion to losses of body
weight for fat (Kotler, 1992). Kotler et al (1989) also
demonstrated that death from wasting in AIDS is related
to the body cell mass depletion rather than the speci®c
underlying cause of the wasting.
The majority of patients in the present study had BMIs
within the normal range, as well as energy intakes within the
range of the RDA, but had insuf®cient intake of micronutrients, including vitamin A, D, C, B6 and niacin, as well as
calcium, iron and zinc. This tendency towards energy and
protein intakes within the ranges of the RDAs with lower
intakes of various micronutrient was also seen in seronegative patients with the same population (N. Silvis & W. F.
Mollentze, 1995; Dannhauser et al, 1996a). Although no
speci®c relationship between micronutrint intake and disease progression could be demonstrated in this study, it is
possible that inadequate micronutrient intake in HIV=AIDS
patients could contribute to disease progression and this
possibility should be investigated further.
Nutrition is a critical determinant of immune response
and malnutrition is the most common cause of immunode®ciency worldwide (Myrvik, 1994; Harbige, 1996; Chandra, 1997). Even a mild state of de®ciency of single
nutrients, such as zinc, selenium, iron, copper, vitamins
A, C, E, B6 and folic acid, may result in an altered immune
response (Chandra, 1997). Furthermore, micronutrient de®ciencies, known to in¯uence immune function, are prevalent even prior to the development of symptoms in HIV
disease (Baum & Shor-Posner, 1998). The inadequate
intake of micronutrients, including vitamins A, C, B6 and
niacin, as well as calcium, iron and zinc, observed in the
present study could have affected the immune status of the
patients. Furthermore, it has been suggested that an imbalance between diminished host antioxidant defences and
increased formation of oxygen radicals and pro-in¯ammatory cytokines create a state of `oxidative stress' in HIV
disease (Look et al, 1997). Together with malnutrition,
including inadequate intake, a multilevel antioxidant de®ciency frequently develops that further heightens oxygen
radical formation. The inadequate intake of antioxidant
nutrients observed in the present study emphasises the
importance of further investigation into the antioxidant
status and effect of supplementation on disease progression
in HIV-infected subjects. However, the role of some
micronutrients, in particular iron and zinc, but also vitamin
A and other antioxidants, in HIV infection seems complex,
and unfavourable effects of high intakes cannot be precluded (Friis & Michaelsen, 1998).
The staple of the population investigated in this study
was traditionally maize-meal porridge, eaten with various
side dishes, including meat and=or vegetable stews or
sauces, according to the socioeconomic status (Bruwer,
1963). With the increase in urbanization, the diets have
become more varied, with increased intakes of bread and
other foods that are bought in the cities and towns, but
maize meal remains the staple food (N. Silvis & W. F.
Mollentze, 1995; Dannhauser et al, 1996a). Although
maize contains appreciable quantities of nicotinic acid,
the major proportion of the substance is present in a form
biologically unavailable to man. Pellagra, resulting from
niacin de®ciency in populations following a maize-based
diet, was identi®ed as an important health problem in South
Africa and forti®cation of maize meal with nicotinic acid
and ribo¯avin has been implemented as a precaution (Du
Plessis, 1983). Ribo¯avin was included in conjunction with
nicotinic acid as symptoms of aribo¯avinoses usually
appear simultaneously with those of pellagra. The levels
of forti®cation were, however, found to be insuf®cient to
eradicate pellagra completely (Agett et al, 1989). Maize is
also de®cient in the amino acid tryptophan, from which the
body is able to produce nicotinic acid. On the other hand,
the median intake of animal protein reported in this study
was more or less 50%, which could probably have supplied
this amino acid. Therefore, although 48% of the subjects in
all three stages of the disease reported inadequate intakes of
niacin, their intakes could have been different depending on
their intake of animal proteins and concentrations of niacin
in the `forti®ed' maize meal. It is thus suggested that the
forti®cation of maize meal in South Africa should be
investigated.
The low intake of vitamins C, A, D, B6, as well as
calcium, iron and zinc by the majority of the subjects could
probably be related to their high intake of maize meal,
which is a poor source of all these nutrients. The relatively
small percentage of subjects with low intakes of magnesium and thiamin could, also be related to the high intake of
maize meal, which is a good source of these nutrients. The
relatively low percentage of subjects with low intakes of
vitamin B12 could possibly be explained by the fact that the
protein intake of the majority of the patients meet the
RDAs, with more or less 50% of the total protein intake
coming from animal sources. The regular consumption of
the green leafy vegetable `marog' (wild spinach), which is
a good source of folate, can possibly explain the low
percentage of subjects with low intakes of this vitamin.
A further aspect that should be kept in mind when
evaluating the nutritional status of HIV-infected patients
in developing countries is the fact that a large number of
them are co-infected with M. tuberculosis. The association
between M. tuberculosis infection and malnutrition is well
documented (Robinson et al, 1986; Niyongabo et al, 1994).
Twenty-six per cent of the patients in this study were coinfected with M. tuberculosis and their dietary intakes as
well as their anthropometric values tended to be lower. The
group III patients infected with M. tuberculosis, however,
tended to have higher intakes than that of the non-infected
group III patients while their anthropometric values tended
to be lower. These differences were not statistically signi®cant. This lack of signi®cance might, however, be due to
the small number of patients in the group. The in¯uence of
171
Nutritional status of HIV-1 positive patients
A Dannhauser et al
172
HIV=M. tuberculosis co-infection on nutritional status
should be investigated further.
In order to study deterioration in nutritional status of
patients as disease progresses, it was attempted to follow
the patients after 2 years. Unfortunately, only a small
number of the original group of patients were available
for the follow-up visits in 1997. Based on this small
number of patients, no signi®cant changes in dietary and
anthropometric pro®les of the patients over the 2-year
period were demonstrated. However, to study the effect
of HIV-associated disease progression on nutritional status,
more complete long-term follow-up studies are needed.
Altogether, these results indicate that this population of
HIV-infected patients is at increased nutritional risk. As the
metabolic abnormalities of HIV-associated diseases have
speci®c causes that are treatable (Kotler et al, 1990; Baum
& Shor-Posner, 1998) and, therefore, potentially reversible,
it is recommended that the bene®t of nutrition invervention
programmes, including supplementation with anti-oxidant
nutrients, as part of the routine care of HIV-infected
patients, be investigated further.
AcknowledgementsÐWe are indebted to G. Joubert for editorial and
statistical advice. We also wish to thank the nursing personnel at the
HIV=AIDS clinic, in particular sisters Lehasa and Jikila, and the Bloemfontein Hospice personnel for help with patient tracing. We are also
indebted to our patients for their willingness to participate.
References
ADSA (Association for Dietectics in Southern Africa) (1996): The role of
nutritional intervention in the treatment of human immunode®ciency
virus infection in adults. S. Afr. J. Foods Sci. Nutr. 8, 162 ± 164.
Aggett N, Van der Westhuyzen, J, Kuyl J & Metz J (1989): Monitoring the
voluntary forti®cation of maize meal with ribo¯avin and nicotinamide.
S Afr. Med. J. 76, 342 ± 343.
Baum MK & Shor-Posner G (1998): Micronutrient status in relationship to
mortality in HIV-1 disease. Nutr. Rev. 56, S135 ± S139.
Baum MK, Shor-Posner G, Lu Y, Rosner B, Sauberlich HE, Fletcher MA,
Szapocznik J, Eisdorfer C, Buring JE & Hennekens CH (1995):
Micronutrients and HIV-1 disease progression. AIDS 9, 1051 ± 1056
(abstract).
Bray GA (1978): De®nition, measurement, and classi®cation of the
syndromes of obesity. Int. J. Obes. 2, 99 ± 112.
Bruwer JP (1963): Die Bantoe van Suid-Afrika, Johannesburg: Hayne &
Gibson.
Castetbon K, Kadio A, Bondurand A, Yao AB, Barouan C, Coulibaly Y,
Anglaret X, Msellati P, Malvy D & Dabis F (1997): Nutritional status
and dietary intakes in human immunode®ciency virus (HIV)-infected
outpatients in Abidjan, CoÃte D'Ivoire, 1995. Eur. J. Clin. Nutr. 51, 81 ±
86.
Chandra RK (1997): Nutrition and the immune system: an introduction.
Am. J. Clin. Nutr. 66, S460S ± S473.
Cuff PA (1990): Acquired immunode®ciency syndrome and malnutrition:
role of gastrointestinal pathology. Nutr. Clin. Pract. 5, 43 ± 53.
Dannhauser A, Bam R, Barnard HC, Badenhorst PN, Slabber M, Badenhorst AMB, Du Toit WC & Joubert G (1996a): Nutritional status with
emphasis on iron and folate intake of pregnant black women with
different levels of iron de®ciency. S. Afr. J. Food Sci. Nutr. 8, 5
(abstract).
Dannhauser A, Joubert G & Nel M (1996b): Nutritional status of preschool
children in the Bloemfontein district. S. Afr. J. Food Sci. Nutr. 8, 14 ± 22.
Dannhauser A, Bester I, Barnard HC, Badenhorst PN, Venter A, Du Toit
E, Slabber M, Badenhorst AM & Nogabe L (1998): Nutritional status
and related factors of preschool-age children in Mangaung. S. Afr. J.
Food Sci. Nutr. 10, (Suppl.), 6 (abstract).
Department of Health, South Africa (1998): Press release, 1998-03-10.
Du Plessis JP (1983): The history of Maize forti®cation in South Africa. S.
Afr. Med. J. 64, 338 ± 339.
Dworkin BM, Wormser GP, Rosenthal WS, Heier SK, Braunstein M,
Weiss L, Jankowski R, Levy D & Weiselberg S (1985): Gastrointestinal
manifestations of the acquired immunode®ciency syndrome: a review of
22 cases. Am. J. Gastroenterol. 80, 774 ± 778.
Dworkin BM, Wormser GP, Axelrod F, Pierre N, Schwardz E, Schwardz E
& Seaton T (1990): Dietary intake in patients with acquired immunode®ciency syndrome (AIDS), patients with AIDS-related complex, and
serologically positive humen immunode®ciency virus patients: correlations with nutritional status. JPEN 14, 605 ± 609.
Eldridge B (1996): Nutritional care in HIV infection and AIDS. In
Krause's Food Nutrition & Diet Therapy, 9th edn & Escott-Stump S,
Mahan LK (ed.) pp 826 ± 994. London: W.B. Saunders.
Feunekes GIJ, Van Staveren WA, De Vries JHM, Burema J & Hautvast
JGAJ (1993): Relative and biomaker-based validity of a food-frequency
questionnaire estimating intake of fats and cholesterol. Am. J. Clin.
Nutr. 58, 489 ± 496.
Fields-Gardner C (1995): A review of mechanisms of wasting in HIV
disease. Nutr. Clin. Pract. 10, 167 ± 176 (abstract).
Food and Nutrition Board. Sub-committee on the Tenth Edition of the
RDAs (1989): Recommended Dietary Allowances, Washington DC:
National Academy Press, pp1 ± 184.
Friis H & Michaelsen KF (1998): Micronutrients in HIV infection: a
review. Eur. J. Clin. Nutr. 52, 157 ± 163.
Frisancho AR (1990): Anthropometric Standards for the Assessment of
Growth and Nutritional Status. Ann Arbor: The University of Michigan
Press.
Grunfeld C, Pang M, Shimizu L, Shigenaga JK, Jensen P & Feingold KR
(1992): Resting energy expenditure, calorie intake, and short-term
weight change in human immunode®ciency virus infection and the
acquired immunode®ciency syndrome. Am. J. Clin. Nutr. 55, 455 ± 460.
Harbige LS (1996): Nutrition and immunity with emphasis on infection
and autoimmune disease. Nutr. Health 10, 285 ± 312 (abstract).
Hecker LM & Kotler DP (1990): Malnutrition in patients with AIDS. Nutr.
Rev. 48, 393 ± 401.
Hogg RS, Zadra JN, Chan-Yan C, Voigt R, Craib KJ, Korosi-Ronco J,
Montaner J & Schechter MT (1995): Analysis of nutritional intake in a
cohort of homosexual men. J. Acquir. Immune De®c. Syndr. Hum.
Retrovirol 9, 162 ± 167.
Jebb SA (1997): Energy metabolism in cancer and immunode®ciency virus
infection. Proc. Nutr. Soc. 56, 763 ± 775.
Kotler PD (1992): Nutritional effects and support in the patient with
acquired immunode®ciency syndrome. J. Nutr. 122, 723 ± 727.
Kotler KP, Wang J & Pierson RN (1985): Body composition studies in
patients with acquired immunode®ciency syndrome. Am. J. Clin. Nutr.
42, 1255 ± 1265.
Kotler DP, Tierney AR, Wang J & Pierson RN (1989): Magnitude of bodycell-mass depletion and the timing of death from wasting in AIDS. Am.
J. Clin. Nutr. 50, 444 ± 447.
Kotler DP, Tierney AR, Brenner SK, Couture S, Wang J & Pierson RN
(1990): Preservation of short-term energy balance in clinically stable
patients with AIDS. Am. J. Clin. Nutr. 51, 7 ± 13.
Lacey CJ, Murphy ME, Sanderson MJ, Monteiro EF, Vail A & Schorah
CH (1996): Antioxidant-micronutrients and HIV infection. Int. J. STD
AIDS 7, 485 ± 489 (abstract).
Langenhoven ML, Conradie PJ, Wolmarans P & Van Eck M (1986):
NRIND Food Quantities Manual, Parow: South African Medial
Research Council pp 1 ± 147.
Langenhoven ML, Kruger M, Gouws E & Faber M (1991): National
Research Institute for Nutritional Diseases (NRIND) of the Medical
Research Council (MRC). Food Composition Tables. 3rd edn. Parow:
South African Medical Research Council pp 1 ± 245
Lee RD & Nieman DC (1993): Anthropometry. In Nutritional Assessment,
ed. RD Lee & DC Nieman. Oxford: WCB Brown & Benchmark pp
121 ± 165.
Leroy V, Salmi LR, Dupon M, Sentilhes A, Texier-Maugein J, Dequae L,
Dabis F & Salamon R (1997): Progression of human immunode®ciency
virus infection in patients with tuberculosis disease. Am. J. Epidemiol
145, 293 ± 300.
Lochner JdeV & Schneider DJ (1994): Die verwantskap tussen tuberkulose, vitamien D, kalsium en VIGS, `n Boodskap vir Suid Africa?. S.
Afr. Med. J. 84, 78 ± 82.
Look MP et al (1997): Serum selenium, plasma glutathione (GSH) and
erythrocyte glutathione peroxidase (GSH-Px)-levels in asymptomatic
versus symptomatic human immunode®ciency virus-1 (HIV-1)-infection. Eur. J. Clin. Nutr. 51, 266 ± 272.
Macallan DC, Noble C, Baldwin C, Fosket M, McManus T & Grif®n GE
(1993): Prospective analysis of patterns of weight change in stage IV
human immunode®ciency virus infection. Am. J. Clin. Nutr. 58, 417 ±
424.
Macallan DC, Noble C, Baldwin C, Jebb SA, Prentice AM, Coward A,
Sawyer MB, McManus TJ & Grif®n GE (1995): Energy expenditure
and wasting in human immunode®ciency virus infection. N. Engl. J.
Med. 333, 83 ± 88.
Nutritional status of HIV-1 positive patients
A Dannhauser et al
Mcorkindale C, Dybevik K, Coulston AM & Sucher KP (1990): Nutritional status of HIV-infected patients during early disease stages. J. Am.
Diet. Assoc. 90, 1236 ± 1241.
Mollentze WF, Moore AJ, Steyn AF, Joubert G, Steyn K, Oosthuizen GM
& Weich DJV (1995): Coronary heart disease risk factors in a rural and
urban Orange Free State black population. S. Afr. Med. J. 85, 90 ± 96.
Morgan D, Maude GH, Malamba SS, Okongo MJ, Wagner H-U, Mulder
DW & Whitworth JA (1997): HIV-1 disease progression and AIDSde®ning disorders in rural Uganda. Lancet 350, 245 ± 250.
Myrvik QN (1994): Immunology and nutrition. In Modern Nutrition in
Health and Disease, 8th edn, ME Shils, JA Olson & M Shike (eds),
Philadelphia: Lea & Fibiger, pp 623 ± 657.
Niyongabo T, Mlika-Cabanne N, Barihuta T, Henzel D, Aubry P. &
Larauze B (1994): Malnutrition, tuberculosis and HIV infection in
Burundi. AIDS 8, 851 ± 852 (letter).
Niyongabo T, Bouchaud O, Henzel D, Melchior JC, Samb B, Dazza MC,
Ruggeri C, Begue JC, Couland JP & Larouze B (1997): Nutritional
status of HIV-seropositive subjects in an AIDS clinic in Paris. Eur. J.
Clin. Nutr. 51, 637 ± 640.
Parisien C, GeÂlinas MD & Cossette M (1993): Comparison of anthropometric measures of men with HIV: asymptomatic, symptomatic, and
AIDS. J. Am. Diet. Assoc. 93, 1404 ± 1408 (abstract).
Paton NIJ, Macallan DC, Jebb SA, Noble C, Baldwin C, Pazianas M &
Grif®n GE (1996): A comparison of methods for measuring changes in
fat-free mass in patients with acquired immune de®ciency syndrome.
Proc. Nutr. Soc. 56, 153A (abstract).
Ralph, A (1993): Dietary reference values. In: Human Nutrition and
Dietetics, 9th edn, JS Gaffow & WPT James, London: Churchill
Livingston, pp 785 ± 796.
Reaidi GB & Cossette M (1992): Vitamin-nutrition issues and AIDS.
Vitamin Information Centre, Medical Update, No. 12.
Resler SS (1988): Nutrition care of AIDS patients. J. Am. Diet. Assoc. 88,
828 ± 832.
Robinson CH, Lawler MR, Chenoweth WL & Garwick AE (1986):
Normal and Therapeutic Nutrition, 17th edn, New York: Macmillan,
pp 475 ± 476.
SAS (1990): SAS Procedures Guide, version 6. 3rd edn, Cary NC: SAS
Institute Inc., pp 991 ± 705
Semba RD (1997): Vitamin A and human immunode®ciency virus infection. Proc. Nutr. Soc. 56, 459 ± 469.
Serwadda D et al (1985): Slim disease: a new disease in Uganda and its
association with HTLV-III infection, Lancet 2, 849 ± 852.
Sharpstone DR et al (1996). Energy balance in asymptomatic HIV
infection AIDS 10 1377 ± 1384 (abstract)
Silvis N & Mollentze WF (1995): Nutritional intake of Black inhabitants
of a semi-rural and a urban area of the Free State Province. (Unpublished)
Skurnick JH, Bogden JD, Baker H, Kemp FW, Sheffet A, Quattrone G &
Louria DB (1996): Micronutrient pro®les in HIV-1-infected heterosexual adults. J. Acquir. Immune De®c. Syndr. Hum. Retrovirol. 12,
75 ± 83.
Summerbell C. (1994): Appetite and nutrition in relation to HIV infection
and AIDS. Proc. Nutr. Soc. 53, 139 ± 150.
Taylor SP & Benatar SR (1989): The Tuberculosis Control ProgrammeÐa
time to re-evaluate? S. Afr. Med. J. 76, 639 ± 640.
Trujillo EB, Borlase BC, Bell S et al (1992): Assessment of nutritional
status, nutrient intake, and nutrition support in AIDS patients. J. Am.
Diet. Assoc. 92, 477 ± 478.
UNAIDS (1998): Report on the Global HIV=AIDS epidemic, Geneva:
WHO=UNAIDS, p 6.
Van der Ryst E, Kotze M, Joubert G, Pieters H, Van der Westhuizen
M, Steyn M, Van Staden AM & Venter C. (1998): Correlation
between total lymphocyte count, absolute CD4‡ count and
CD4‡ percentage in a group of HIV-infected South African
patients. J. Acquir. Immune De®c. Syndr. Hum Retrovirol 19: 238 ±
244.
Van Staden AM, Barnard HC, Nel M, Attwood EM, Dannhauser A, Van
der Ryst E & Le Roux GD. Nutritional status of HIV-seropositive
patients in the Free State Province of South Africa: laboratory parameters Centri. Afr. J. Med. (in press).
Wilkinson D & Davies GR (1997): The increasing burden of tuberculosis
in rural South AfricaÐimpact on the HIV epidemic. S. Afr. Med. J. 87,
447 ± 450.
Wilkinson D & Moore DAJ (1996): HIV-related tuberculosis in
South AfricaÐclinical features and outcome. S. Afr. Med. J. 86,
64 ± 67.
173