Feeding the emergence of advanced heart disease in Soweto: a nutritional survey of black African patients with heart failure.

Abstract

To describe dietary habits and potential nutritional deficiencies in black African patients diagnosed with heart failure (HF).
Dietary intake in 50 consecutively consenting HF patients (mean age: 47 ± 18 years, 54% female) attending a major hospital in Soweto, South Africa were surveyed using validated quantitative food frequency questionnaires. Food intakes, translated into nutrient data were compared with recommended values. In women, food choices likely to negatively impact on heart health included added sugar [consumed by 75%: median daily intake (interquartile range) 16 g (10-20)], sweet drinks [54%: 310 ml (85-400)] and salted snacks [61%: 15 g (2-17)]. Corresponding figures for men were added sugar [74%: 15 g (10-15)], sweet drinks [65%: 439 ml (71-670)] and salted snacks [74%: 15 g (4-22)]. The womens' intake of calcium, vitamin C and vitamin E was only 66, 37 and 40% of the age-specific requirement, respectively. For men, equivalent figures were 66, 87 and 67%. Mean sodium intake was 2 372 g/day for men and 1 972 g/day for women, 470 and 294% respectively, of recommended consumption levels.
The nutritional status of black African patients with HF could be improved by recommending healthier food choices and by reducing the intake of sweet drinks and excess salt.

Full text

Page 1

CARDIOVASCULAR JOURNAL OF AFRICA • Advance Online Publication, July 2011
AFRICA
1
Cardiovascular Topics
Feeding the emergence of advanced heart disease in
Soweto: a nutritional survey of black African patients
with heart failure
SANDRA PRETORIUS, KAREN SLIWA, VERENA RUF, KAREN WALKER, SIMON STEWART
Summary
Aim: To describe dietary habits and potential nutritional
deficiencies in black African patients diagnosed with heart
failure (HF).
Methods and Results: Dietary intake in 50 consecutively
consenting HF patients (mean age: 47 ± 18 years, 54%
female) attending a major hospital in Soweto, South Africa
were surveyed using validated quantitative food frequency
questionnaires. Food intakes, translated into nutrient data
were compared with recommended values. In women, food
choices likely to negatively impact on heart health includ-
ed added sugar [consumed by 75%: median daily intake
(interquartile range) 16 g (10–20)], sweet drinks [54%:
310 ml (85–400)] and salted snacks [61%: 15 g (2–17)].
Corresponding figures for men were added sugar [74%: 15
g (10–15)], sweet drinks [65%: 439 ml (71–670)] and salted
snacks [74%: 15 g (4–22)]. The womens’ intake of calcium,
vitamin C and vitamin E was only 66, 37 and 40% of the
age-specific requirement, respectively. For men, equivalent
figures were 66, 87 and 67%. Mean sodium intake was 2 372
g/day for men and 1 972 g/day for women, 470 and 294%
respectively, of recommended consumption levels.
Conclusions: The nutritional status of black African patients
with HF could be improved by recommending healthier food
choices and by reducing the intake of sweet drinks and excess
salt.
Keywords: heart failure, Africa, food preferences, malnutrition,
salt
Submitted 17/9/10, accepted 21/5/11
Cardiovasc J Afr 2011; 22: online publication www.cvja.co.za
DOI: CVJ-21.100
Heart failure (HF) has become a major public health problem in
that, unlike other cardiovascular diseases, the number of people
discharged from hospital with a diagnosis of HF is increasing.1
In developed countries, HF can be observed in 2 to 3% of the
population and asymptomatic ventricular dysfunction is evident
in about 4% of the population.2 This will increase with age, and
in the 70- to 80-year age group, HF can be observed in between
10 and 20% of people.2,3 The epidemic of cardiovascular disease
(CVD) has probably stabilised in developed countries, but devel-
oping countries are increasingly suffering from the emerging
burden of CVD.4
As populations in South Africa and sub-Saharan Africa under-
go economic development, the disease profile shifts and CVD
becomes a growing cause of death and disability.5 Previously
considered a rarity in Africa and predominantly caused by
infectious disease or idiopathic dilated cardiomyopathy,6 the
syndrome of heart failure (HF) has emerged as a challenging
public health problem in sub-Saharan Africa.7
The Heart of Soweto (HOS) study8 has documented a much
higher-than-expected burden of modifiable risk factors9 and
advanced forms of heart disease10 linked to epidemiological
transition in one of Africa’s largest urban concentrations of black
Africans. Data from the HOS study showed that during the peri-
od from 2006 to 2008, of the 5 328 de novo cases captured with
heart disease, 2 505 (47%) of these cases presented with chronic
heart failure.11,12 Ominously, in addition to the ‘traditional’ causes
of HF in Africa, such as idiopathic dilated cardiomyopathy, rheu-
matic fever, HIV-related cardiomyopathy, peripartum cardiomyo-
pathy and hypertensive heart failure, ‘lifestyle’ factors, including
hypertension, obesity, dyslipidaemia and type 2 diabetes mellitis
(particularly in women), appear to have expanded the pathways
to, and burden of, HF in this community.8
Although the natural history of HF in Africa is still differ-
ent from that of high-income countries, it results in the same
high level of preventable morbidity and premature mortal-
ity.13,14 In those countries already in the midst of an epidemic
of HF, multidisciplinary management programmes targeting
the common factors leading to clinical instability have been
successfully developed.15-17 Certain positive measures have been
implemented in low- and middle-income countries for disease
prevention, including WHO initiatives.5 However, inadequate
funding hinders efforts to establish adequate multidisciplinary
management programmes targeting the common factors leading
to heart disease in South Africa.5 Moreover, the role of dieticians
has been largely confined to patients with lipid disorders, obesi-
Soweto Cardiovascular Research Unit, Department of
Cardiology, Chris Hani Baragwanath Hospital, University of
the Witwatersrand, Johannesburg, South Africa
Sandra Pretorius, RD (SA), sandra.pretorius@nashuaisp.co.za
Karen Sliwa, MD, PhD, FESC
Verena Ruf, MD
Baker IDI Heart and Diabetes Institute and Monash
University, Melbourne, Australia
Karen Walker, PhD, MND (APD)
Monash University, Melbourne, Australia
Simon Stewart, PhD, FCSANZ, NFESC, FAHA

Page 2

CARDIOVASCULAR JOURNAL OF AFRICA • Advance Online Publication, July 2011
2
AFRICA
ty, diabetes and/renal failure.18 If, however, nutritional education
and promotion of good nutrition could be better understood and
recognised to be inclusive of behavioural change, then it will be
viewed as a necessary component within contemporary cardiac
rehabilitation and self-management programmes.18
One cornerstone of HF management particularly relevant
to urban Africans affected by HF is dietary modification. For
example, sodium restriction (2–3 g/day) is standard therapy in
the management of symptomatic chronic HF, and black individu-
als are particularly responsive to this strategy.19,20 However, lack
of adherence and poor self-care behaviours persist, with dietary
indiscretions contributing to a substantial portion (up to 20%)
of hospital readmissions.21 Specific dietary interventions play an
important role in improving health outcomes.21,22
Three major studies addressing food choices and dietary
patterns in adult black South Africans were identified from the
literature. However, given the historical rarity of the syndrome,
there are very little data to describe the dietary habits of
specifically urban African patients with HF. The Dikgale study23
examined food choices, nutrient intake and weight status of
black adults. The Transition, Health and Urbanisation study
(THUSA)24 examined the food choices, health status and the
effect of urbanisation on a black population. The Black Risk
Factor study (BRISK)25 examined the risk factors for developing
CVD in urban black Africans. Data from these studies show that
rural black adults have a very low consumption of fat and a high
consumption of carbohydrates, typical of the traditional rural
African diet.23,25
Urbanisation is associated with markedly increased intake
of fat, sugar, meat and beverages.23-25 Although a decrease in the
consumption of maize porridge with urbanisation was found,
it is still consumed in high amounts by these black population
groups.23 As the traditional diet is abandoned in favour of a
Western diet, food choices shift away from complex carbohy-
drates and higher fibre to foods high in fat, bringing an increased
risk for chronic diseases of lifestyle.26 According to Stewart et
al., data on the population of Soweto have shown a low preva-
lence for CVD and the underlying risk factors.8 This situation
however may be changing, as urbanisation and the nutritional
transition in South Africa is accompanied by an increase in the
CVD risk factors in black Africans.27
The overall study aim was therefore to provide a detailed
description of the dietary habits and potential nutritional defi-
ciencies in a subgroup of urban black African patients diagnosed
with HF, living in Soweto, South Africa, and managed via the
Cardiology Unit of the Chris Hani Baragwanath Hospital. It
focused on the impact of varied dietary patterns, the poor socio-
economic status of many patients and probable lack of awareness
of the contribution of poor nutrition to cardiovascular disease.
Ultimately, these data will be used to identify key targets for
more culturally sensitive support and to argue for a greater role
for dieticians in the management of an increasing number of
urban black South Africans affected by HF.12
Methods
As part of the previously described Heart of Soweto study,8
detailed demographic and clinical data are captured from all
individuals with heart disease presenting to the Chris Hani
Baragwanath Hospital, Soweto, via a prospective clinical regis-
try. In 2006, this included 1 960 patients presenting with a prima-
ry or secondary diagnosis of HF (an average of 162 patients per
month). All were diagnosed by echocardiography and specialist
cardiological review. This was a prospectively planned study
of 50 consecutively consenting black Africans (28 females, 22
males), referred to the Heart Failure Clinic in 2006/7 with a
documented diagnosis of HF.
The study was approved by the Human Research Ethics
Committee (Medical), University of the Witwatersrand,
Johannesburg, M050550. All participants provided written
informed consent. The study fully conformed to the principles
outlined in the Declaration of Helsinki.
Dietary instrument, data and nutrient analyses and
recommendations
In addition to the detailed clinical and demographic data
collected as part of the Heart of Soweto Clinical Registry, an
interviewer-administered quantitative food frequency question-
naire (QFFQ) was collected at a point in time when patients
had received either limited, or no instructions for a low-sodium,
low-fat therapeutic diet for HF.28,29 A quantitative food frequency
questionnaire is a validated questionnaire to determine food
choices and consumption. The previously validated QFFQ
used in this study was developed by a researcher at Northwest
University. This QFFQ has previously been used to evaluate the
food choices of the African population living in the North West
Province, South Africa, as part of the THUSA study.30,31
The quantitative QFFQ has been validated via statistical
methods in an African population.30 It includes 139 types of
food and records how often a given type of food is consumed
as: time/s per day, per week, per month. It also records prepara-
tion methods. Quantities of food eaten were determined in rela-
tion to pictures of standardised portions of the most commonly
consumed foods (e.g. maize meal porridge, rice, meat, etc.). The
researcher also used standardised cups, teaspoons etc. to estimate
portion sizes. The patients were also asked to name foods eaten
that were not included in the questionnaire and to point out ques-
tions that were unclear or difficult to understand.
The QFFQ was administered through interview by the
researcher, SP, who is a registered dietician in the Heart Failure
Outpatient Clinic, Chris Hani Baragwanath Hospital, and trained
in administering the QFFQ.
Food data were translated into nutrient data using the Medical
Research Council (MRC) Food Finder 3, 2007, which is based
on South African food composition tables. Total dietary starch
was calculated from the total amount of carbohydrates minus
the sum of total dietary fibre plus added sugars. To assess the
consumption of high-sodium foods, data were aggregated to
provide percentages of high-sodium foods consumed both daily
and weekly.
Collated dietary patterns and nutritional intake data were
compared to the South African Food-Based Dietary Guidelines.
Importantly, one guideline advises that unrefined or minimally
processed starchy foods, such as maize, wheat, sorghum, oats,
rice in the form of porridges, breads, pastas, samp, breakfast
cereals and other products should be the main food around which
the rest of the meal is planned.32 Promotion of carbohydrate-rich
foods contributes to optimal nutrient intake, particularly in low-
income groups. Largely unrefined carbohydrate-rich foods are

Page 3

CARDIOVASCULAR JOURNAL OF AFRICA • Advance Online Publication, July 2011
AFRICA
3
excellent sources of dietary fibre and provide several important
vitamins and minerals.33 It is also recommended that in a healthy,
balanced diet, protective against chronic diseases of lifestyle, at
least 55% of the total energy (%E) should be provided by a vari-
ety of carbohydrate-rich foods, with around 30%E provided by
fat and 15%E by protein. To provide at least 55%E in an 8 500
kJ diet, at least 275 g carbohydrate should be consumed daily.32
Demographic and clinical data
At the time the QFFQ was administered, body mass was meas-
ured with an electronic digital scale, measuring up to 200 kg in
graduations of 0.1 kg (Seca 767), and body height was taken
with a telescopic measuring rod (Seca 220) attached to the scale,
to the nearest 1 mm. Data on the clinical [including left ventricu-
lar ejection fraction, New York Heart Association (NYHA) func-
tional class and concurrent diagnoses] and socio-demographic
profile (including age, gender and educational status) were
collected prospectively.
Statistical analyses
Data were analysed using SPSS for Windows version 14.0.1
(SPSS Inc, Chicago, Illinois). Normally and non-normally
distributed continuous data are given as the mean (standard
deviation: SD) and median (interquartile range: IQR), respec-
tively. Categorical data are presented as counts and percentages.
Proportional data were compared via the Chi-squared test while
all nutrient data were compared via the Mann Whitney U-test
according to gender, and actual versus recommended dietary
intake. Significance has been accepted as p < 0.05 (two-tailed).
Results
The demographic and clinical profile of the study cohort is
shown in Table 1. Reflective of the overall Heart of Soweto
study cohort, there were more women (56%) than men. Women
were slightly, but not significantly younger than the men and the
entire HF cohort was typically two decades younger than that
seen in high-income countries. Hypertension and obesity were
highly prevalent in both genders. Concurrent diabetes was also
common, particularly in men. The majority of patients had left
ventricular systolic dysfunction (LVEF < 45%) and symptoms
of exercise intolerance and dyspnoea indicative of NYHA func-
tional class II or III.
The daily food consumption of the cohort as measured by the
QFFQ according to gender is shown in Table 2. Significantly,
more women (79%) than men (65%) reported eating brown or
wholemeal bread, 75% of women and 48% of men consumed
sweets and chocolates, processed meat were eaten by 89% of
women and 78% of men, and packet soup was consumed by 57%
of women and 43% of men. Women, but few men also reported
eating stock cubes, 18% and 4% respectively, possibly since
TABLE 1. DEMOGRAPHIC AND CLINICAL
PROFILE OF THE STUDY COHORT
Socio-demographic profile
Mean age (years)1
No education
1–5 years’ education
6–10 years’ education
Post-matriculation qualifications
Risk profile
Body mass index (kg/m2)1
Hypertension
Diabetes
Heart failure profile
NYHA class II
NYHA class III
NYHA class IV
Left ventricular ejection fraction1
Men
n = 22 (%)
51 ± 12
1 (4.5)
5 (23)
15 (68)
1 (4.5)
Women
n = 28 (%)
47 ± 18
4 (14)
7 (25)
16 (57)
1 (3.6)
25.2 ± 4.8
14 (65)
2 (10)
26.5 ± 6.4
18 (65)
2 (7.6)
11 (50)
4 (19)
0 (0)
37.3 ± 9.1%
12 (43)
9 (32)
1 (3.6)
36.4 ± 13.4%
1Data are given as mean ± SD or as number (%).
TABLE 2. DAILY FOOD CONSUMPTION OF HF PATIENTS
ACCORDING TO
THE QUANTITATIVE FOOD FREQUENCY QUESTIONNAIRE
Men (n = 22)
Median
daily intake
(interquartile
Women (n = 28)
Foods/food groups
Maize meal (g)
Mabella (g)1
Oats (g)
Potatoes (g)
White bread (g)
Brown/wholegrain
bread (g)
Cereals: refined (g)
Cereals:
wholegrain (g)
Mageu (ml)2
Added sugar (g)
Sweets and
chocolates (g)
Cakes and biscuits (g)
Cold drinks
(sweetened) (ml)
Meat, chicken, fish,
eggs (g)
Milk and milk
products (ml)
Legumes (g)
Fruit (fresh) (g)
Vegetables (fresh) (g)
Margarine on
bread (g)
Salt added to cooked
food (g)
Salted snacks (g)
Take-away foods (g)
Sauces and
condiments
Stock cubes
Packet soup
Processed meat
Propor-
tion
(%)
91
52
26
78
22
65
range)
Propor-
tion
(%)
93
57
32
86*
29
79***
Median
daily intake
(interquartile
range)
424 (140–688)
111 (55–136)
80 (50–100)
59 (28–89)
73 (38–98)
87 (60–113)
516 (200–750)
78 (25–64)
88 (55–107)
76 (28–91)
88 (50–60)
102 (43–120)
13
22
13 (13–15)
29 (25–30)
14
21
7 (4–10)
17 (9–15)
30
74
48
208 (43–321)
15 (10–15)
19 (7–30)
39
75
64 (16–71)
16 (10–20)
11 (3–12)75***
48
65
45(15–25)
439 (71–670)
57
54*
7(5–10)*
310 (85–400)
100 150 (105–190) 100127 (83–168)
87262 (129–370)93*113 (58-145)*
43
100
100
83
18 (10–24)
174 (150–160)
76 (40–103)
15 (7–20)
43
100
100
75
18 (9–28)
147 (40–160)
78 (50–91)
16 (10–20)
91 2 (2–2) 75***2 (2–2)
74
48
57
15 (4–22)
23 (10–15)
7 (2–10)
61**
32***
64
15 (2–17)
16 (10–25)
4 (2–5)
4
43
78
1 (1–1)
3 (1–5)
35 (8–54)
18***
57**
89**
1 (1–2)
2 (1–2)
26 (8–35)
1Unrefined porridge made from sorghum; 2dried and broken corn kernels;
3a carbohydrate-rich drink made from fermented mealie (maize) meal and
malt.
Significant difference between men and women, *p < 0.05, **p < 0.01 ,
***p < 0.001.

Page 4

CARDIOVASCULAR JOURNAL OF AFRICA • Advance Online Publication, July 2011
4
AFRICA
the women were more aware that they were added during food
preparation. Conversely, more men than women reported the
consumption of takeaway foods, 48 and 32%, respectively. Salted
snacks were eaten by 74% of men and 61% of women, while
more men than women reported adding salt to cooked food, 91%
and 75% respectively. Despite differences in the proportions of
men and women selecting certain foods, the median daily intake
of foods eaten was broadly similar for men and women. Specific
differences included a higher median intake for men of milk and
milk products, and for cakes and biscuits (both p < 0.05).
Median daily nutrient intake in this HF population group
is shown in Table 3. Although men consumed a significantly
greater quantity of protein (p < 0.05), protein as a percentage
of energy was similar (around 13%E) for both men and women.
Both men and women consumed high amounts of carbohydrate
(47–52%E). Although added sugar intake was low (< 10%E),
fibre intake was moderately low, suggesting that many carbohy-
drate foods eaten came from refined sources, rather than from
wholegrain cereals, as recommended. Both women and men
consumed < 30%E from fat. Consumption of saturated fat and
trans fat was significantly lower in women than men (p < 0.05, p
= 0.001, respectively). Four men (18%) consumed alcohol, with
one reporting consumption equivalent to 27 g per day. Only one
woman (3.5%) drank alcohol.
Table 4 indicates mean daily micronutrient intake for men and
women. In men, the mean intake of calcium and magnesium and
of vitamins C, D, E and folate was inadequate. Mean intakes of
these nutrients were also inadequate in women although mean
intakes of vitamin D and magnesium were only marginally low.
The mean intake of iron in women was only 50% of the level
recommended, while mean intakes of riboflavin, vitamin B6,
pantothenate and niacin were also moderately low.
Fig. 1 indicates that often over half of this patient group had
inadequate micronutrient consumption, while all the women and
the majority of men consumed excessive amounts of sodium.
Sodium intake was 470% above recommended intake levels in
men and 294% above recommended intake levels in women.
As seen in Fig. 2, most sodium came from bread and processed
foods. In the body, the ratio of sodium (in extracellular fluid) to
potassium (in intracellular fluid) is about 2:3. As seen in Table 4,
the intake of potassium in relation to sodium was too low, due to
the increased consumption of processed food and the inadequate
intake of fruits, vegetables and unrefined cereals.
The likely cost of consuming a healthy diet in Soweto was
calculated based on food prices relative to minimum income
support available in May 2008. Current food intake required
an expenditure of approximately 40% of the current disability
grant, which in 2008 was R940 per month. A recommended
food intake, where maize meal porridge is supplemented with
mabella (coarse), legumes, carrots, spinach, apples, oranges and
full-cream milk would require an expenditure of only 30% of this
benefit and therefore represents an attractive option both from a
financial and health status perspective.
Discussion
The most significant finding is the inadequate nutrient intake and
excessive salt consumption in this high-risk HF patient cohort.
Processed and convenience foods contributed to the high intake
of salt as well saturated and trans fatty acids. Low consumption
of fruit and vegetables contributed to the low micronutrient and
dietary fibre intake. Overall, the pattern of dietary consumption
observed is likely to have been a major contributor to the pattern
of sub-optimal health outcomes (i.e. premature mortality and
recurrent morbid events) found in these patients from Soweto
TABLE 3. ENERGY AND DAILY NUTRIENT INTAKE
OF HF PATIENTS BASED ON A QUANTITATIVE
FOOD FREQUENCY QUESTIONNAIRE
Nutrient
Energy (kJ)
Protein (g)
% plant-derived
% total energy
Total carbohydrate (g)
% total energy
Starch (g)
Dietary fibre (g)
Added sugars (g)
Total fat (g)
% total energy
Saturated fat (g)
% total energy
Monounsaturated fat (g)
% total energy
Polyunsaturated fat (g)
% total energy
Total trans fat (g)
Cholesterol (mg)
Significant difference between men and women, *p < 0.05, **p = 0.001.
Daily intake [median (interquartile range)]
Men (n = 22)
9 145 (6 857–12 879)
74.0 (101–62)
42.8
13.8
272 (223–404)
47.6
17.8 (13–27)
20.6 (14–25)
40.2 (23–76)
65.7 (5–91)
26.6
19.9 (16–29)
8.1
22.9 (17–31)
9.3
15.7 (11–22)
6.3
0.94 (0.62–1.8)
308 (177–403)
Women (n = 28)
7 472 (5 568–9 478)
58.8 (51–66)*
43.5
13.4
245 (170–336)
52.5
11.0 (8–17)
16.2 (13–23)
33.1 (19–69)
47.4 (39–81)
23.5
15.1 (12–20)*
7.5
17.0 (14–26)
8.4
12.7 (9–25)
6.3
0.46 (0.28–0.67)**
214 (160–307)
TABLE 4. MICRONUTRIENT INTAKE OF HF PATIENTS IN
RELATION TO RECOMMENDED DIETARY INTAKES
Daily
intake
men
Vitamin D (mcg) 4.5
Vitamin C (mg)78
Magnesium (mg) 361
Vitamin E (mcg) 10
Calcium (mg) 655
Folate (mcg)227
Iron (mg)
Riboflavin (mg)
Vitamin B6 (mg)
Pantothenate (mg)
Niacin (mg)
Potassium (mg)
Adequate intake
Sodium (mg)2.372
+1 872 (470)
Potassium (mg) 2512
+0.512 (150)
Vitamin B12 (mcg)6.3
Pantothenate (mg)6.5
Biotin (mcg) 39
Iron (mg) 11
Riboflavin (mg)1.5
Niacin (mg)18
Vitamin B6 (mg)1.4
Thiamine (mg) 1.3
Vitamin A (RE) (mcg) 949
Micronutrient
deficiency
Difference
from DRI
(%)
–0.5 (90)
–12 (87)
–59 (86)
–5 (67)
–345 (66)
–173 (57)
Daily
intake
women
4.7
47
292
9
411
187
9
1.0
1.2
4.6
13.4
1938
Difference
from DRI
(%)
–0.3 (6)
–28 (37)
–29 (9)
–6 (40)
–789 (66)
–213 (53)
–9 (50)
–0.1 (9)
–0.1 (8)
–0.4 (8)
–0.6 (4)
–0.62 (4)
1 972
+1 472 (294)
+3.9 (260)
+1.5 (130)
+9.0 (130)
+3 (125)
+0.2 (115)
+2.0 (113)
+0.1 (108)
+0.1 (108)
+49 (105)
6.1
+3.7 (254)
34
+4 (13)
1.1
970
0 (0)
+270 (39)

Page 5

CARDIOVASCULAR JOURNAL OF AFRICA • Advance Online Publication, July 2011
AFRICA
5
with HF.8
High salt intake, particularly in men, was a major problem in
this black urban patient group. This, related to a high consump-
tion of bread, processed and take-away foods and the use of high-
salt stock cubes and sauces, consistent with North American
findings where salt in bread and pre-prepared and cereal foods
contributed to around one-quarter of total salt intake.33 Possible
barriers to adherence to a healthy, low-salt diet in this black
population were: lack of knowledge regarding high-salt foods
and healthy affordable alternatives, perceptions that meals
prepared without added salt were tasteless and boring, and lack
of support for dietary change from family members.34
Although a salt restriction (2–3 g/day) is standard therapy
for HF,35 black Sowetans with HF commonly consumed 5–7 g
per day.36 This indicates the need for higher levels of dietetic
education to achieve sodium-restricted diets. At Chris Hani
Baragwanath Hospital, 10 registered dieticians currently provide
a nutritional service to 2 500 patients; clearly an inadequate ratio
of 1:250, instead of the more acceptable ratio of 1:50.37
In contrast to the rural areas of South Africa where more ‘tradi-
tional’ food patterns still apply, in the urban areas undergoing
very rapid epidemiological transition, poor quality ‘Westernised’
diets are common.26,38 The South African Dietary Guideline
(SADG) addresses these nutritional issues, although compliance
with recommendations is not readily achieved by disadvantaged
urban populations.38 The SADG for example recommend serv-
ings of ‘meat, fish, chicken or eggs’ should be eaten daily as
nutrient-rich sources of high-quality protein. As selection of
fatty meats and full-fat dairy foods can increase cardiovascular
disease risk, Scholtz and colleagues39 suggest a safe daily intake
would comprise: 400–500 ml milk, two to three servings of fish
and four eggs, and no more than 560 g of meat per week.
In this group of CHF patients, median intake is less than
half this amount, presumably as these foods are not afford-
able. Nevertheless, the proportion of dietary protein was within
accepted levels (13%E), although the majority came from plant
rather than animal sources, with implications for micronutrient
intake. Calcium intake, particularly in women was inadequate.
Some more affordable sources of plant protein, notably legumes,
rich in many nutrients, were not selected in quantity, suggesting
lack of familiarity with preparing meals using these foods.
The total fat intake seen in this HF cohort was within recom-
mended levels (< 30%E) but saturated fat intake was excessive,
particularly in men, and was related to choice of poor-quality
fatty meat, high-fat dairy foods, cakes and biscuits, and take-
away foods. This is consistent with the trend for higher total fat
and saturated fat consumption seen with urbanisation through-
out South Africa.40 This patient group continued to eat more
traditional carbohydrate foods such as maize porridge, oats and
mabella, but also consumed highly refined carbohydrate foods
such as cakes, biscuits, cold drinks, sweets, chocolates and added
sugar, liable to increase triglycerides and to promote insulin
resistance and obesity.33
Fruit and vegetables provide alternative sources of carbohy-
drates and contain many cardioprotective nutrients,41 including
potassium (lowers blood pressure), folate (reduces plasma homo-
Fig. 1. Proportion of men and women consuming more than the recommended daily intake of sodium or with less
than the daily recommended intake of selected micronutrients. Significant difference between men and women, *p <
0.01, **p < 0.001.
100
80
60
40
20
0
Vit DVit CVit ECa FolateIron Vit B6Niacin Sodium
Percent
Men
Women
90%
6%
87%
37%
67%
40%
66% 66%
57%
53%
125%
50%
108%
8%
113%
4%
470% 294%
Fig. 2. Percentage of men and women consuming the 10 foods contributing most to mean sodium intake in HF patients.
100
80
60
40
20
0
Salt added
to cooked
food
Processed
meat
MargarineSauces &
condiments
Takeaway
foods
Packet
soup
BreadSalted
snacks
Stock
cubes
Breakfast
cereal
Percent
Men
Women