Prevalence of iron and zinc deficiencies among preschool children ages 3 to 5 y in Vhembe district, Limpopo province, South Africa

Accepted Manuscript Prevalence of Iron and Zinc deficiencies among preschool children aged 3-5 years in Vhembe district, Limpopo province, South Africa S.A. Motadi, X.G. Mbhenyane, H.V. Mbhatsani, N.S. Mabapa, R.L. Mamabolo PII:

S0899-9007(14)00438-9

DOI:

10.1016/j.nut.2014.09.016

Reference:

NUT 9382

To appear in:

Nutrition

Received Date: 11 June 2014 Revised Date:

18 August 2014

Accepted Date: 9 September 2014

Please cite this article as: Motadi S, Mbhenyane X, Mbhatsani H, Mabapa N, Mamabolo R, Prevalence of Iron and Zinc deficiencies among preschool children aged 3-5 years in Vhembe district, Limpopo province, South Africa, Nutrition (2014), doi: 10.1016/j.nut.2014.09.016. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

1

Prevalence of Iron and Zinc deficiencies among preschool children aged 3-5

2

years in Vhembe district, Limpopo province, South Africa

3

1

4 5

1

6 7

2

Motadi SA, 2Mbhenyane XG, 1Mbhatsani HV, 1Mabapa NS, 1Mamabolo RL

RI PT

Nutrition Department, School of Health Sciences, University of Venda, Thohoyandou, 0950, South Africa Department of Interdisciplinary Health Sciences, Faculty of Medicine and Health Sciences, Stellenbosch University, Maitland, 7602, Stellenbosch, South Africa

9

SC

8

ABSTRACT

Background: Children under five years constitute the most vulnerable group and their nutritional status

11

is a sensitive indicator of community health and nutrition.

12

Objective: To determine the prevalence of zinc and iron deficiency among preschool children aged 3-5

13

years in Vhembe district, Limpopo province, South Africa.

14

Design: This study included 349 preschool children recruited from two municipalities of Vhembe

15

district, Limpopo province, South Africa. Municipalities were purposively selected and subjects were

16

chosen by simple random sampling. Body weight and height were measured using standard techniques.

17

Serum zinc, iron, ferritin, transferrin saturation, transferrin and CRP levels were also assesses as well as

18

haemoglobin levels.

19

Results: The prevalence of wasting, stunting and underweight was 1.4%, 18.6% and 0.3% respectively

20

while 20.9% of the children were overweight and 9.7% were obese. The prevalence of zinc deficiency

21

was 42.6% and anaemia was 28%, and both were higher in female as compared to male children. When

22

using serum ferritin and transferrin saturation 7 (2%) children had iron deficiency anaemia. Combined

23

iron and zinc deficiencies using ferritin was found in 8 (2.3%) of the children while when using

24

transferrin saturation they were found in 42 (12%) children.

25

Conclusion: Iron and zinc deficiencies as well as anaemia were common in the preschool children

26

accompanied by high prevalence of stunting; and overweight and obesity. The results call for

27

interventions to combat this escalating problem of child malnutrition in the form of nutritional education

28

for mothers and food handlers at the preschools so as to ensure food diversification in these children.

AC C

EP

TE D

M AN U

10

29 30

KEYWORDS: Iron deficiency, zinc deficiency, pre-school children, stunting, overweight 1

ACCEPTED MANUSCRIPT

31

INTRODUCTION

32

Despite the legislated fortification of staple foods in South Africa [1], there is a general shortage

34

of iron and zinc in a large part of the population’s diets [2]. Furthermore, iron and zinc

35

deficiencies still appear to be important public health problems in many low–income countries

36

[2], as iron deficiency has an impact on psychomotor development, cognitive functions and

37

growth [3]. On the other hand zinc deficiency has been associated with poor growth, depressed

38

immune function, increased susceptibility to and severity of infections and neurobehavioral

39

abnormalities [3]. Diets of many people in rural communities of South Africa consist of plant

40

sources with very minimal amounts of animal sources which are rich sources of both iron and

41

zinc [4].

M AN U

SC

RI PT

33

42

Iron deficiency is particularly prevalent in infants and young children. These age groups have a

44

higher risk of developing iron deficiency because of the high demand of iron during the period of

45

rapid growth. Iron deficiency exists in all countries, but its magnitude is higher in South East

46

Asia (57%) and Africa (46%) [5]. The cause of nutritional anaemia may be due to inadequate

47

iron intake from food and low iron bioavailability.

48

TE D

43

Recently, there has been heightened interest in the role of zinc in humans, as it plays an

50

indispensable role as a key component of a host of enzymes crucial for optimal metabolism and

51

body functioning [6]. In addition, zinc is an anti-inflammatory and antioxidant agent and also

52

functions in cell-mediated immune responses, while in childhood it is required for optimal

53

growth and development, as well as brain functioning [7].

AC C

54

EP

49

55

In developing countries, for instance, poor nutrition (especially micronutrient deficiencies) is not

56

only responsible for over 10 million preventable child deaths yearly [8], it is also a risk factor

57

preventing an additional estimated 200 million children under 5 years old from attaining their

58

full cognitive developmental potential [3]. Like many other developing populations, a recent

59

national survey conducted among South African children revealed that 27.9% were anemic while

60

45.3% were zinc deficient [9]. Although large nutritional national studies [5, 9, 10] have been

61

conducted among South African children, there is paucity of data concerning the prevalence of 2

ACCEPTED MANUSCRIPT

62

iron and zinc deficiency in rural pre-school children in particular. In view of this the study was

63

conducted to evaluate iron and zinc status in preschool children residing in a rural area of

64

Limpopo province, South Africa.

65

SUBJECTS AND METHODS

RI PT

66 67 68

Study population

69

The study was conducted in Vhembe district which is one of five municipal districts of Limpopo

71

province. The district is divided into four local municipalities namely Thulamela, Makhado,

72

Musina and Mutale.

M AN U

SC

70

73

Simple random sampling was used to select two municipalities (Mutale and Makhado). Mutale

75

had an estimated population of 91870 and 148 villages whereas Makhado had an estimated

76

population of 516,031 and 290 villages [11]. A list of preschools was obtained from the

77

Department of Education, Vhembe district, and each preschool was assigned a number from

78

which 16 were randomly selected to be included in the study. The same procedure was used to

79

select 380 children (3-5 years of age) to participate in the study. Children who were 3-5 years old

80

and whose parents consented and were present on the day of data collection were included in the

81

study. The children had to be free from infection and not taking any medication and present on

82

the day of data collection. Children whose parents/guardians did not give consent were excluded

83

from the study.

EP

AC C

84

TE D

74

85

Anthropometric assessments were performed according to standard procedures described by the

86

International Society for the Advancement of Kinanthropometry [12]. The following

87

measurements were taken in duplicate using calibrated equipments with the children wearing

88

light clothing and no shoes: standing height and weight. Height was measured to the nearest 0.1

89

cm using a calibrated portable stadiometer; weight was measured to the nearest 0.01 kg on a

90

portable Seca solar scale (model 0213). The solar scale and stadiometer were calibrated before

91

measurements using a calibration weight and steel tape, respectively.

92

3

ACCEPTED MANUSCRIPT

A professional nurse checked for clinical signs of zinc deficiency by administering an oral zinc

94

taste test. This was done by placing 10ml of liquid zinc in the child’s mouth, then letting the

95

child hold it in the mouth for ten to fifteen seconds while observing facial expressions and taste.

96

The nurse then asked the children what liquid zinc tasted like and recorded the results. Clinical

97

signs which could be identified by the naked eye were also recorded.

RI PT

93

98 99

Biochemical analysis

100

Fasting venous blood was collected for blood analysis. Haemoglobin was measured by

102

standardized procedures at Ampath laboratories using a STKS analyzer (Beckman Coulter Inc.,

103

California, USA), using 3-level controls provided by the manufacturer, within 2h of blood

104

collection. Serum zinc was measured using

105

immunosorbent assays were used to measure serum transferrin saturation (Tsaturation) and

106

ferritin (Ramco Laboratories, Inc., Stafford, Texas) and C-reactive protein (CRP) (Human

107

Biochemical and Diagnostic Laboratories, South Africa) by SYNCHRON LX® System(s),

108

UniCel® DxC 600/800 System(s) and SYNCHRON® Systems CAL 5

109

immunoturbidimetric test (Human Biochemical and Diagnostic Laboratories, South Africa).

110

The coefficients of variance (CV) for all assays were <10%.

M AN U

125

SC

101

I- Radioimmunoassay Enzyme-linked

TE D

Plus by

111

Definitions of Iron and Zinc Deficiencies and Anaemia

113

Anaemia was defined as haemoglobin levels <11 g/dl for the children and iron deficiency as

114

SF<12 µg/l or % transferrin saturation <15% [13]. Haemoglobin concentrations less than 7.0g/dl

115

were considered severe anaemia, 7.0 to 9.9 g/dl as moderate anaemia and Hb> 10.0 g/dl to <

116

11.0 g/dl as mild anaemia [14], while Zinc deficiency was defined as serum zinc concentration <

117

9.9 µmol/dl [6]. Iron deficiency anaemia was defined as having low haemoglobin levels

118

accompanied by low transferrin saturation or serum ferritin or both low serum ferritin and

119

Tsaturation levels.

AC C

EP

112

120 121 122 4

ACCEPTED MANUSCRIPT

123

Ethical considerations

124

Ethical clearance was obtained from the University of Venda Research Ethics Committee

126

(SHS/12/NUT/02) and the study was approved by the Provincial Department of Health Research

127

Committee and the Department of Education. The study was performed in accordance with

128

principles of the Declaration of Helsinki (2008), Good Clinical Practices and the laws of South

129

Africa. An oral and written explanation of the study, including possible risks, was given to the

130

parents and guardians. Parents and caregivers gave written signed consent for their children to

131

participate in the study while the children gave verbal assent.

SC

132

Statistical Methods

M AN U

133

RI PT

125

134

The children’s weight and height were entered into WHO Anthro-plus version 1.0.2 software

136

(http://www.who.into/nutgrowthdb/into, Accessed 22/03/2013) for the calculation of weight for

137

age (WAZ), height-for-age (HAZ), weight-for-height (WHZ) and body mass index-for-age

138

(BAZ)z-scores. The data was then exported into Statistical Package for Social Sciences (SPSS)

139

version 21 for further analysis together with the biochemical measurements. Descriptive statistics

140

were computed on the data and the median and interquartile ranges were used to describe

141

continuous data as the data was not normally distributed while frequencies were used to describe

142

categorical data. Spearman’s correlation coefficients were computed to compare relationships

143

between anthropometric measurements and biochemical measures. For comparison of groups

144

the Mann-Whitney U test was used for continuous data and Χ2- test for categorical data. Logistic

145

regressions were computed to assess biochemical predictors of nutritional status (HAZ, WAZ,

146

WHZ and BAZ).

148 149

EP

AC C

147

TE D

135

150 151 152 153

5

ACCEPTED MANUSCRIPT

154

RESULTS

155

Three hundred and eighty children were enrolled into the study, but the number later reduced to

157

349 (92% response rate) due to non-response from the parents/guardians or the children being

158

absent during the day of data collection. The median age of the children was 4.22 years (4.63

159

years in boys and 4.54 years in girls). There were no gender differences between the

160

anthropometric and biochemical measurements of the children.

SC

161

RI PT

156

Table 1 shows the median and interquartile ranges of anthropometric and biochemical

163

measurements between the stunted and non-stunted children. The non-stunted children were

164

heavier than the stunted children (p<0.0001), while weight-for-age z score values for non-stunted

165

children were higher than those of the stunted children (p<0.0001). BMI for age z-score was

166

higher in the stunted versus the non-stunted children (p<0.0001). There were no differences in

167

biochemical measurements between the two groups.

M AN U

162

169 170 171

TE D

168

Table 1: Median and Interquartile Range of anthropometric and biochemical measurements of stunted and non-stunted children (n= 349).

The prevalence of wasting, stunting and underweight was 1.4%, 18.6% and 0.3% respectively

173

while 20.9% of the children were overweight and 9.7% were obese. Stunting was more prevalent

174

in boys (21.5%) as compared to girls (16.1%) though not statistically significant. Furthermore,

175

3.1% of boys were wasted while no girls were found to be wasted (Χ2=5.788; p=0.016). In boys

176

0.6% were underweight while no girls were found to be underweight. Overweight and obesity

177

levels were virtually the same in both boys (20.9% and 9.8% respectively) and girls (20.9% and

178

9.7% respectively) (Table 2).

AC C

EP

172

179 180

Table 2: Nutritional status of the children as defined by their anthropometric indices

6

ACCEPTED MANUSCRIPT

Ferritin levels below 12µg/dl were taken as indicative of iron deficiency and it was found that 25

182

(7.2%) of the children were iron deficient. Of the female children 14 (7.5%) had low serum

183

ferritin as compared to 11 (6.7%) of the boys. Furthermore, 98 (28%) of the children were iron

184

deficient when transferrin saturation below 15% was taken as an indicator of iron deficiency. In

185

addition, 53 (28.5%) girls had low transferrin saturation as compared to 45 (27.6%) boys (Table

186

3). Using both serum ferritin and transferrin saturation as markers of iron deficiency revealed

187

that 15 (4.2%) of the children were iron deficient.

SC

188

RI PT

181

The prevalence of anaemia was higher in girls (29.6%) as compared to boys (26.4%), though not

190

statistically significant. When iron deficiency anaemia (IDA) was determined by using

191

transferrin saturation as a marker of iron deficiency this indicated that 36 (10.3%) children had

192

IDA, while when using serum ferritin levels it was found that 12 (3.4%) children had IDA.

193

Furthermore, when using both serum ferritin and Tsaturation levels as markers of iron deficiency

194

7 (2%) children were found to have IDA.

TE D

195

M AN U

189

Haemoglobin concentrations less than 7.0g/dl were considered severe anaemia, 7.0 to 9.9 g/dl as

197

moderate anaemia and Hb> 10.0 g/dl to < 11.0 g/dl as mild anaemia [14]. Of all the children 14

198

(4%) had moderate anaemia while 84 (24%) had mild anaemia. In boys 4.3% had moderate

199

anaemia as compared to 3.7% of girls. Furthermore, 36 (22.1%) boys had mild anaemia as

200

compared to 48 (25.8%) girls.

EP

196

AC C

201 202

Zinc deficiency was defined as serum zinc concentration < 9.9 µmol/dl [6]. Of all the children 149

203

(42.6%) were zinc deficient and the prevalence of zinc deficiency was higher in girls (45.2%) as

204

compared to boys (39.9%) (Table 3). Combined Iron and zinc deficiencies using ferritin were

205

found in 8 (2.3%) of the children while when using Tsaturation it was found that 42 (12%)

206

children had combined iron and zinc deficiencies. Furthermore, when iron status was assessed

207

using both serum ferritin and Tsaturation it was found that 6 (1.7%) had combined iron and zinc

208

deficiencies. 7

ACCEPTED MANUSCRIPT

209 210

Table 3: Prevalence of micronutrient deficiencies as determined by the selected biochemical variables

211

Bad taste of liquid zinc was considered to be indicative of good serum levels of zinc while those

213

who indicated that they had little taste of sweetness were considered to have low serum zinc

214

levels. Children who indicated that liquid zinc was tasteless were considered to be extremely

215

zinc deficient [15]. Fourteen children (4%) indicated that liquid zinc tasted bad while 105 (30%)

216

indicated that there was little taste than sweet. Furthermore, 230 (65.9%) indicated that liquid

217

zinc was tasteless. In girls 8 (4.3%) indicated that liquid zinc tasted bad as compared to 6 (3.7%)

218

boys. Furthermore, 60 (32.3%) girls indicated that they had little taste than sweet as compared to

219

45 (27.6%) boys. However, 112 (68.7%) boys indicated that liquid zinc was tasteless as

220

compared to 118 (63.4%) girls. In addition 0.5% of the children showed symptoms of zinc

221

deficiency and the prevalence was the same in both gender groups (Table 4).

M AN U

SC

RI PT

212

222

224

Table 4: Zinc Taste Test results of the children by gender

TE D

223

In the total population CRP was negatively correlated with transferrin (r= -0.110; p=0.040) and

226

Tsaturation (r= -0.277; p<0.0001) but positively correlated with ferritin (r= 0.128; p= 0.017).

227

Zinc was positively correlated with Tsaturation (r=0.107; p=0.045) but negatively with CRP (r= -

228

0.112; p=0.037). Furthermore, CRP was negatively correlated with weight (r= -0.156; p= 0.003).

229

WHZ was positively correlated with transferrin (r=0.169; p=0.002) but negatively correlated

230

with zinc (r=-0.174; p=0.001). HAZ was negatively correlated with CRP (r=-0.145; p=0.007)

231

and zinc (r=-0.122; p=0.022). WAZ was positively correlated with transferrin (r= 0.136;

232

p=0.038) while BAZ showed a positive correlation with transferrin (r=0.24; p= 0.021) (Table 5).

AC C

EP

225

233

234

Table 5: Correlation between biochemical markers and anthropometric measurements

235

8

ACCEPTED MANUSCRIPT

In girls CRP was negatively correlated with transferrin (r=-0.124; p=0.091) and Tsaturation (r= -

237

0.237; p<0.0001). WHZ showed a negative correlation with serum ferritin (r= -0.161; p=0.028)

238

and zinc (r= -0.194; p= 0.008). HAZ was negatively correlated with CRP (r= -0.146; P =0.046),

239

and BAZ was negatively correlated with zinc (r=-0.164; p=0.025). In boys CRP correlated

240

negatively with Tsaturation (r= -0.325; p<0.0001). Zinc was positively correlated with

241

Tsaturation (r= 0.225; p=0.004) but negatively correlated with CRP (r= -0.188; p= 0.016).

242

Transferrin was positively correlated with WHZ (r= 0.2; p= 0.011) and WAZ (r= 0.249;

243

p=0.011). CRP was negatively correlated with weight (r= -0.202; p= 0.010), WHZ (r=-0.155;

244

p=0.048) and WAZ (r= -0.198; p=0.045) (data not shown).

SC

M AN U

245

RI PT

236

In stunted children CRP was negatively correlated with Tsaturation (r= -0.292; p=0.018) (data

247

not shown). While in non-stunted children CRP was negatively correlated with transferrin (r= -

248

0.133; p=0.025) and Tsaturation (r=-0.266; p<0.0001). Serum zinc was positively correlated with

249

Tsaturation (r= 0.140; p= 0.018). Transferrin was positively correlated with WHZ (r=0.175;

250

p=0.003), WAZ (r=0.156; P= 0.033) and BAZ (r=0.165; p=0.005). Tsaturation was negatively

251

correlated with WAZ (r=-0.161; p=0.027). Ferritin was negatively correlated with WHZ (r=-

252

0.148; p=0.013). Furthermore, CRP was negatively correlated with HAZ (r= -0.137; p=0.021),

253

and Zinc was negatively correlated with WHZ (r= -0.216; p<0.0001), WAZ (r= -0.145; p=0.048)

254

and BAZ (r= -0.190; p<0.0001) (Table 6).

258

EP

256 257

Table 6: Correlations between biochemical markers and anthropometric measurements in non-stunted children

AC C

255

TE D

246

259

Selected biochemical measures were fitted into a stepwise linear regression model to see which

260

of them predicted the nutritional status of the children (Table 7). The predictors of stunting were

261

CRP and Zinc levels while the predictors of wasting were the child’s age and zinc and

262

overweight was predicted only by age.

263

9

ACCEPTED MANUSCRIPT

264 265

Table 7: Linear regression models for assessing selected biochemical predictors of malnutrition states in children

266

DISCUSSION

268

Pre-school children are at an important stage of life where nutrition plays an important role in

269

growth and development and has long lasting effects in later life. Children under five years

270

constitute the most vulnerable group of any community and their nutritional status is a sensitive

271

indicator of community health and nutrition. There were no gender differences observed in both

272

biochemical and anthropometric measurements in the current study. These findings are

273

congruent with previous studies which indicated that there were no significant gender differences

274

in serum iron and zinc concentrations at this tender age [16]. In addition a study done by

275

Wamani et al. on children aged 0-5 years also indicated no gender differences in anthropometric

276

measurements [17]. This in itself suggests that at this young age there might be several common

277

factors between the genders which act together to mask any differences in growth rate and

278

nutritional requirements.

M AN U

SC

RI PT

267

TE D

279

The prevalence of wasting, stunting and underweight was 1.4%, 18.6% and 0.3% respectively

281

while 20.9% of the children were overweight and 9.7% were obese. The nutritional status of the

282

current pre-school children might have been affected by the access that the community,

283

household and individuals had on food and utilization of the available food. Parents and

284

caregivers make most food choices for meals consumed at home and these food choices are

285

based on culture, beliefs, cost, time constraints and availability which may result in poor intake

286

of vital nutrients necessary for growth [18]. In addition, lack of knowledge on nutritional needs

287

of children, inadequate food intake, poor feeding practices, low income, poor weaning and

288

complementary feeding practices have been shown to contribute to inadequate energy and

289

protein intake leading to malnutrition [19]. Unfortunately, these were not measured in this study

290

as it would have given a better insight into the problem faced by these children. Socio-economic

291

disparities, poor nutritional qualities of traditional diets as well as insufficient food intake are

292

however, considered the main causes of malnutrition in many rural communities [19].

AC C

EP

280

293 10

ACCEPTED MANUSCRIPT

Undernutrition and overnutrition coexisted in the current study like in many rural communities

295

of South Africa [20]. Interestingly in many communities where there was a high prevalence of

296

undernutrition there was also a high prevalence of overweight and obesity [21]. Historically, a

297

heavy child was associated with being “healthy” but today this perception has drastically

298

changed on the basis of evidence that obesity and overweight in childhood are associated with a

299

wide range of serious health complications and increased risk of premature illness and death later

300

in life [22].

RI PT

294

301

Baughcum et al. reported that a high percentage of mothers with a poor education did not

303

consider their overweight children to be overweight [23]. It is believed that mothers have their

304

own culturally-based and influenced definitions of obesity [24]. Thus with the cultural belief that

305

“bigger is healthier” [18, 23] in the groups studied, the mother of a lean healthy baby might feel

306

that her child was not getting sufficient food and as such, change her feeding practices. As

307

opposed to adults, where obesity is an immediate risk factor for most chronic diseases, in

308

children it rarely poses an immediate and serious health threat. In effect, its short term

309

consequences are often psychosocial and include outcomes such as low self-esteem, poor peer

310

acceptance and low participation in social and sporting activities [22]. Available evidence

311

suggests that the influence of childhood obesity on adult status rises with age, with a low

312

predictive value in children below 2 years, a significant predictor at 3-5 years, and an even

313

stronger one at older ages [25].

M AN U

TE D

EP

314

SC

302

Compared to what was reported by Kruger et al. iron deficiency in the current study varied

316

wherein the findings for serum ferritin were lower (7.2% v/s 15.6%) and those for transferrin

317

saturation were higher (28.1% v/s 11%) [26]. In addition it was found that depending on the iron

318

marker used combined iron and zinc deficiencies ranged between 2.3% and 12%. It has been

319

reported that though ferritin has a high specificity as a marker of iron status, its levels are

320

affected by the presence of infections and inflammation as it is an acute-phase reactant [27].

321

Transferrin saturation levels on the other hand are known to be a less sensitive marker of iron

322

status as they may decrease before anaemia develops but not early enough to can identify iron

323

depletion and they are affected by malnutrition, inflammation and chronic infections [27]. As

324

such the differences in the prevalences when using the various biochemical markers were used,

AC C

315

11

ACCEPTED MANUSCRIPT

could have resulted from these factors. Furthermore, the presence of parasitic infections such as

326

hookworm, ascaris, amoeba cyst, strongyloides, schistosoma, and malaria may have resulted in

327

the observed low serum iron, Tsaturation and increased ferritin [28]. These blood depleting

328

parasites have been reported to suck the blood voraciously and damage the mucosa while also

329

competing for iron with the child’s body [3]. In addition it has been previously shown that in

330

Vhembe district there was a high prevalence of infections in both adults and children [29, 30,

331

31]. Unfortunately deworming before the study was not done as this may have avoided the

332

competition for nutrients in these children as such revealing a clearer picture of their

333

micronutrient status.

SC

RI PT

325

334

This high prevalence of anaemia observed in these children (28%) may be due to the fact that

336

children consume less food than adults and their diet often consists of foods with low iron

337

content and in which iron bioavailability is low. This low proportion of foods rich in iron has

338

also been reported from the menus offered by pre-schools in South Africa [32]. Unfortunately,

339

pre-school menus were not analysed as they would have given a better understanding of the

340

children’s dietary intake. However, it has also been reported that iron rich foods were poorly

341

consumed in rural communities of South Africa because of their high cost and in some cases

342

cultural or religious practices [4]. Even though South Africa had legislated fortification of flour

343

and maize to reduce micronutrients malnutrition [1], food introduced during childhood includes

344

small amounts of food rich in iron with a high content of substances such as phytates and fibre

345

that may interfere with iron absorption [33].

TE D

EP

346

M AN U

335

Transferrin saturation correlated negatively with WAZ while ferritin correlated negatively with

348

WHZ and CRP correlated negatively with height. However, transferrin was positively correlated

349

with WHZ, WAZ and BAZ suggesting a relationship between body composition and iron status

350

a finding which has been reported in adults [34]. Iron requirements in growing overweight

351

children have been found to be higher than in their normal weight counterparts. Thus, at a similar

352

level of iron intake the iron status of overweight children deteriorates to a greater extent than in

353

the normal weight children during periods of growth [34]. Overweight children have also been

354

shown to have higher concentrations of CRP, interleukin-6, leptin and hepcidin that may predict

355

poor iron status [35]. Despite this, it has been indicated that iron deficiency does not appear to

AC C

347

12

ACCEPTED MANUSCRIPT

356

contribute to growth faltering [36], but inadequate intake and low content of bioavailable iron are

357

known to have a direct impact on nutritional status of young children leading to anaemia in

358

underweight, wasted and stunted children [6].

359

The prevalence of zinc deficiency was 42.6% with no gender differences but this signifies a point

361

of concern as to what might be facing this next generation especially with regard to the vicious

362

circle of malnutrition [37] as zinc is required for growth and development in children [8]. These

363

findings are congruent with those of the South African national food consumption survey

364

fortification baseline-1; where it was found that 45.3% of children aged 1-9 years had inadequate

365

zinc intakes and, therefore, were considered to be at risk of zinc deficiency. However, the

366

prevalence was higher in Limpopo province (27.3%) [38]. As such these results warrant

367

immediate action since the prevalence of zinc deficiency was higher than 20% which is regarded

368

as a threshold to raise public health concerns [6]. The most probable reason for these findings

369

may be attributed to the fact that local villagers plough maize and take it to milling stations

370

where there is no fortification [2].

371

M AN U

SC

RI PT

360

Zinc deficiency is associated with lack of gustin which leads to low salivary zinc concentration

373

and can potentially lead to impaired taste acuity and loss of appetite [39]. The prevalence of zinc

374

deficiency using liquid zinc taste test was 30%, which was higher in girls as compared to boys.

375

In addition 0.5% of the children showed clinical signs of zinc deficiency. The symptoms of zinc

376

deficiency might not appear immediately in childhood but have greater consequences later in life

377

and furthermore, the prevalence was lower as compared to the prevalence as determined by

378

serum zinc. The reason for the differences might be that the study participants were children who

379

at this age are considered to be unable to express differences in taste perception and a reliable

380

estimation was achieved by using the child’s own spontaneous verbal judgement [39].

381

Furthermore, due to several factors such as oral hygiene children might express different facial

382

responses even when they have the same zinc levels. Zinc Taste Test (ZTT) can be useful when

383

determining zinc status since impaired taste acuity has also been associated with suboptimal zinc

384

status in children [40].

AC C

EP

TE D

372

385

13

ACCEPTED MANUSCRIPT

It has been shown that during childhood zinc deficiency contributes to stunting and impaired

387

cognitive development as well as increased prevalence of infectious diarrhoea, pneumonia and

388

malaria [6]. Furthermore, it has been reported that diets rich in plant-based foods result in poor

389

bioavailability of zinc, owing to their high levels of phytates which are known to form

390

complexes with zinc that render it un-absorbable [41]. On the other hand, Manary et al. observed

391

that reducing phytates intake in children did not necessarily improve zinc absorption. This in

392

itself shows the multi-linkages in nutrients during absorption [42]. As such it is reasonable to

393

speculate that with insufficient intakes of several essential minerals the children’s metabolism

394

may adapt by transporting the small amounts consumed to other crucial organs at the expense of

395

linear growth.

SC

RI PT

386

M AN U

396

Zinc correlated positively with WHZ, WAZ and BAZ. Previous studies have assessed growth

398

velocity associated with modest dietary supplements and have confirmed the effect of zinc

399

supplementation in increasing height and weight [43, 44]. Furthermore, in the current study

400

congruent with these previous findings zinc predicted 11.3% of the variations in wasting together

401

with age while obesity was predicted by age only though to a lesser extend (1.3%). Daily intake

402

of zinc is required especially during childhood because the body has no specialised zinc storage

403

system. Varying traditional diets, differences in dietary intake of zinc in some areas along with

404

different soil zinc levels may be other reasons that could result in these negative associations

405

[16]. The observed results for both iron and zinc deficiencies though warrant more studies that

406

will also be examining the dietary intake of these pre-school children, which unfortunately were

407

not measured in the current study. This will help in elucidating the cause of the problems that

408

they experience in order to secure a healthy future for them.

EP

AC C

409

TE D

397

410

It has been suggested that stunting was a functional indicator of population zinc status [43]

411

findings similar to the current study where serum zinc was found to be a predictor of stunting

412

accounting for 2% of the variations. There was a high prevalence of both zinc deficiency

413

(42.6%) and stunting (18.6%) observed in the present study as such suggesting that zinc

414

deficiency may be a contributing factor to the stunting levels observed but on the other hand it is

415

also known that socioeconomic factors and genetics are other underlying causes because of their

416

indirect influence on the health and nutrition of the population [45]. These other relationships 14

ACCEPTED MANUSCRIPT

417

should be considered when studying zinc deficiency and stunting, as it has been shown that

418

stunting results from insufficient feeding, inadequate care and nutrient depletion caused by

419

repeated bouts of illness, depressed appetite and poverty [45].

420

Insufficient intakes of these two micronutrients from their rich sources may lead to growth

422

faltering, delayed development and contribute to morbidity and mortality from diarrhoea,

423

pneumonia and stunting in children from developing countries [8, 43]. Iron and zinc have been

424

found to interact either at the site of absorption or post absorption because they compete for

425

similar transport pathways making one to be more prevalent than the other [46]. They compete

426

for absorption or interact at each other’s site of absorption; therefore, the benefit of

427

supplementing both may be less than if either was given alone [47]. Sandstrom et al found that

428

increasing the molar ratio of iron to zinc from 1:1 to 2.5:1 did not affect zinc absorption. An iron

429

and zinc ratio of 25:1 decreased zinc absorption but addition of the zinc ligand histidine

430

decreased the inhibitory effect of this dose of iron and zinc absorption which increased from

431

34% to 47% [48]. Furthermore, they also observed no inhibitory effect on zinc absorption when

432

iron and zinc were given with a meal [48]. Iron and zinc deficiencies coexist in malnourished

433

children in developing nations [47]. The inverse relationship between iron and zinc in male

434

children supports this existing knowledge of iron and zinc interaction either at the site of

435

absorption or post absorption because they compete for similar transport pathways making one to

436

be more prevalent than the other [46].

SC

M AN U

TE D

EP

437

RI PT

421

Lack of employment and education among mothers and guardians in rural areas are often the

439

effects of under development and are primary causes of poverty and this could result in poor

440

food choices that could have reduced inclusion of iron and zinc rich foods in the children’s diet

441

[46]. Educational level and socioeconomic status of the households were not measured in the

442

current study and this might have given an insight into other possible factors responsible for the

443

high prevalence of iron and zinc deficiencies as well as stunting; and overweight and obesity.

AC C

438

444 445

Inadequate intakes of iron and zinc have also been identified among segments of South African

446

infants due to improper weaning practices and children due to poor feeding practices [49]. Poor 15

ACCEPTED MANUSCRIPT

qualities of traditional diets accompanied by infectious diseases as well as insufficient food

448

intake were considered the main causes of malnutrition in many populations [6]. The type of diet

449

consumed, which is predominantly cereal and vegetable-based, contains absorption inhibitors

450

such as phytates, dietary fibre, oxalates, tannins and phenols [34]. Tatala et al. reported that even

451

though iron might be available in the cereal based diets its bioavailability is low when measured

452

by in vitro methods [50].

RI PT

447

453

Many low income households in rural areas of South Africa are highly dependent for survival on

455

government grants of which the old age and child grants are the most important [11] and this

456

makes it difficult for them to purchase food which could meet their daily requirements of

457

nutrients because of the high cost of foods especially the protein rich food. The high cost of

458

animal products makes these rich sources of absorbable iron and zinc inaccessible to many of the

459

world’s poorer populations [6]. In the case of 1-5year-old children in Limpopo province [51] per

460

capita intake of meat and chicken was only 40.5 g/day. The per capita intake of green leafy

461

vegetables was 48.8 g/day. However, the latter would have contributed mainly non-haem iron,

462

which is not very bioavailable.

TE D

463

M AN U

SC

454

The low prevalence of underweight, wasting and high prevalence of both stunting; and

465

overweight and obesity still exists among children in rural areas. The prevalence of zinc

466

deficiency may be considered a public health problem among pre-schoolers in Vhembe district.

467

The high prevalence of anaemia on the other hand could be as a result of the poor sources and

468

consequently the low bioavailability of iron in the diet. Improving nutritional status of these

469

children should follow an integrated approach of tackling both malnutrition and micronutrients

470

deficiencies at the same time considering behavioural approaches. Nutrition education

471

programmes targeted at mothers and guardians on the importance of iron and zinc in the early

472

development of children should include the importance of dietary diversification of meals

473

(especially the inclusion of animal products) for pre-schoolers and improved food

474

processing/handling in order to assist in alleviating this problem. This could be done by ensuring

475

that government put in place mechanisms that will result in the introduction on nutrition

476

education and proper feeding practices during their pre-natal visits and the messages being

477

emphasized again when they visit the health centres for their post-natal visits. In addition pre-

AC C

EP

464

16

ACCEPTED MANUSCRIPT

schools can be allocated nutrition professionals who would assist them in making sure that they

479

offer cost-effective healthy meals which will assist in boosting the nutritional status of the

480

children. In addition families especially mothers and the community at large need to be

481

sensitized on the issues around healthy eating and its importance for their children’s

482

development and educational achievements. Based on this future studies looking at the effects of

483

these interventions are warranted to assess their feasibility in the current population whilst taking

484

into consideration their cultural and religious beliefs.

485

Acknowledgements

SC

486

RI PT

478

487

The authors would like to acknowledge the University of Venda and the South African National

489

Research Foundation for their financial assistance towards conducting the study. Furthermore we

490

would like to pass our message of appreciation to the mothers/guardians of the children, their

491

children and staff members at the preschools for their participation and cooperation, not

492

forgetting the nursing staff which assisted during blood collection. Any opinions, findings and

493

conclusions or recommendations expressed in this article are those of the authors and the NRF

494

does not accept any liability in regard thereto.

495 496

Conflict of Interest

497

500 501 502 503 504

EP

499

None declared.

AC C

498

TE D

M AN U

488

505 506 507 508

17

ACCEPTED MANUSCRIPT

509

References

510

[1] National Food Fortification Task Group. Pretoria: South Africa Department of Health; 2002.

512

[2] Dhansay MA, Marais CD, Labadarios D. Zinc status. In: Labadarios D. National food

513

consumption survey: fortification baseline, Chapter 9D: South Africa 2005. Stellenbosch:

514

Department of Health 2007.

515

[3] Grantham-McGregor S, Cheung YB, Cueto S, Glewwe Pl, Richter L, Strupp B, et al.

516

Developmental potential in the first 5 years for children in developing countries. Lancet, 2007;

517

369: 60-70.

518

[4] Labadarios D, Swart R, Maunder EMW, Kruger HS, Gericke GJ, Kuzawayo PMM.

519

‘Executive summary of the National Food Consumption Survey Fortification Baseline (NFCS-

520

FB-I), South Africa, 2008; SA J Clin Nutr, 2005; 21(3, suppl. 2): 245–300.

521

[5] World Health Organization. Iron deficiency anemia; assessment, prevention and control: a

522

guide for programmer managers, Geneva; 2001.

523

[6] International Zinc Nutrition Consultative (IZiNCG). Quantifying the risk zinc deficiency,

524

Technical Brief No 1; 2007.

525

[7] Gibson RS. Zinc: The missing link in combating micronutrient malnutrition in developing

526

countries. Proceedings of the Nutrition Society, 2006; 65(1): 51-60.

527

[8] Black RE. Zinc deficiency, infectious disease and mortality in the developing world. J Nutr,

528

2003; 133: 1485S-1489S.

529

[9] Labadarios D, Swart R, Maunder EMW, Kruger HS, Gericke GJ, Kuzawayo PMM. Folate,

530

iron, vitamin A and zinc status of children 12-108 months and women 16-35 years old in South

531

Africa: the national food consumption survey-fortification Baseline. Abstract 70. 2008 Nutrition

532

congress: Evidence based Nutrition Leading the Way in Innovation, South Africa p202; 2008.

533

[10] Samuel FO. Egal AA, Oldewagen-Theron WH, Napier CE,Venter CS, 2010.Prevalence of

534

zinc deficiency among primary school children in poor peri-urban informal settlement in South

535

Africa. Institute of Sustainable Livelihoods, Vaal University of Technology, South Africa.

536

[11] Statistics South Africa. 2011. Pretoria. 1 vol.

537

[12] Stewart A, Marfell-Jones M, Olds T, de Ridder H International Standards for

538

Anthropometric Assessments (ISAK). National Library of Australia 2011; pp.114.

AC C

EP

TE D

M AN U

SC

RI PT

511

18

ACCEPTED MANUSCRIPT

539

[13]. Worldwide prevalence of anaemia 1993–2005: WHO global database on anaemia / Edited

540

by

541

http://whqlibdoc.who.int/publications/2008/9789241596657_eng.pdf.

542

[14] World Health Organization. Preventing and controlling iron deficiency anaemia through

543

primary health care: a guide for health administrators and programme. nep published by:

544

Kathmandu: Development Communication Consultancy Services 58 p; 1989.

545

[15] Crosby. Zinc Diagnostic Tests 2009; http//:www.diagnoseme.com.

546

[16] Dehghani SM, Katibeh P, Haghighat M, Moravej H, Asadi S. prevalence of zinc deficiency

547

in 3-18 years old children in Shiraz-Iran. Iran Red Crescent Med J, 2011; 13(1): 4-8.

548

[17] Wamani H, Astrøm AN, Peterson S, Tumwine JK, Tylleskar T. Boys are more stunted than

549

girls in Sub-Saharan Africa: a meta-analysis of 16 demographic and health surveys. BMC

550

Pediatrics, 2007; 7: 17.

551

[18] Kruger R, Gericke GJ. A qualitative exploration of rural feeding and weaning practices,

552

knowledge and attitudes on nutrition. Public Health Nutr, 2003; 6: 217-223.

553

[19] Mittal PC, Srivastava S. Diet, nutritional status and food related traditions of Oraon tribes of

554

New Mal (West Bengal), India. Rural Remote Health, 2006; 6: 385. Available from:

555

http://www.rrh.org.au

556

[20] Zere E, McIntyre D .Equity in self-reported adult illness and use of health service in South

557

Africa: inter-temporal comparison. J Health Popul Nutr, 2003; 21: 205-215.

558

[21] Kazemi TF. Prevalence of Overweight and Obesity in 7 to 18 Year-Old Children in

559

Birjand/ Iran. Iranian J Pediatrics, 2009; 19(2): 135-140.

560

[22] Dietz WH. Health consequences of obesity in youth: childhood predictors of adult disease.

561

Pediatrics 1998; 101: 518-525.

562

[23] Baughcum AE, Chamberlin LA, Deeks CM, Powers SW, Whitaker RC. Maternal

563

perceptions of overweight preschool children. Pediatrics, 2000; 106: 1380-1385.

564

[24] Baughcum AE, Powers SW, Johnson SB, et al. Maternal feeding practices and beliefs and

565

their relationships to overweight in early childhood. J Dev Behav Pediatr 2001, 22: 391-408.

566

[25] Whitaker RC, Wright JA, Pepe MS, Siedel KD, Dietz WH. Predicting obesity in young

567

adulthood from childhood and parental obesity. N Engl J Med 1997; 337: 869-873.

de

Benoist,

Erin

McLean,

Ines

Egli

and

Mary

Cogswell.

AC C

EP

TE D

M AN U

SC

RI PT

Bruno

19

ACCEPTED MANUSCRIPT

[26] Kruger M, Badenhorst C J, Mansvelt EPG, Laubscher JA, Benade AJS. Effects of iron

569

fortification in a school feeding scheme and anthelmintic therapy on the iron status and growth

570

of six- to eight-year-old school children. Food Nutr. Bull 1996; 17: 11–21.

571

[27] Wu AC, Lesperance L and Bernstein H. Screening for iron deficiency. Pediatrics in Review,

572

2002; 23(5): 171-177.

573

[28] Stoltzfus R, Chwaya H, Montresor A, Albonico M, Savioli L and Tielsch J. Malaria,

574

hookworms and recent fever related to anemia and iron status indicators in 0- to 5-y Zanzibari

575

children and these relationships change with age. J Nutr, 2000; 130: 1724–1733.

576

[29] Momoh A, Davies TC, Akinsola HA, Iweriabor B, Mashao MB, Geophagy, anaemia and

577

geohelminth infection amongst non-pregnant women in Vhembe district, South Africa Int Jnl

578

Cur Res., 2012; 4(11): 129-134.

579

[30] Samie A, Guerrant RL, Barrett L, Bessong PO, Igumbor EO and Obi CL. Prevalence of

580

intestinal parasitic and Bacterial Pathogens in diarrhoeal and Non-diarrhoeal human stools from

581

Vhembe district South Africa. J Health Popul Nutr., 2009; 27 (6): 739-745.

582

[31] Samie A, Nchachi DJ, Obi CL, Igumbor EO, Prevalence and temporal distribution of

583

Schistosoma haematobium infections in the Vhembe district, Limpopo Province, South Africa.

584

Afr J Biotechnol., 2010; 9 (42): 7157-7164.

585

[32] Vieria F, Ana Claudia, Diniz AS, Cabral CP, Oliveira RS, Lola MMF, et al. Nutritional

586

assessment of iron status and anemia in children under 5 years old at public day care centers. J

587

Pediatr (Rio J), 2007; 83(4): 370-376.

588

[33] Cole CR, Grant FK, Swaby-Ellis ED, Smith JL, Jacques A, Northrop-Clewes C, et al. Zinc

589

and iron deficiency and their interrelations in low-income African American and Hispanic

590

children in Atlanta. Am J Clin Nutr., 2010; 91(4): 1027–1034.

591

[34] Baumgartner J, Smuts CM, Aleberli I, Malan L, Tjalsma, Zimmermann MB. Overweight

592

impairs efficacy of iron supplementation in iron-deficient South African children: a randomized

593

controlled intervention. Int J Obesity, 2013; 37: 24-30.

594

[35] Aeberli I, Hurrell RF, Zimmermann MB. Overweight children have higher circulating

595

hepcidin concentrations and lower iron status but have dietary iron intakes and bioavailability

596

comparable to normal weight children. Int J Obesity, 2009; 33: 1111–1117.

AC C

EP

TE D

M AN U

SC

RI PT

568

20

ACCEPTED MANUSCRIPT

[36] Torbjorn L, Lonnerdal Bo, Stenlund H, Indria LG, Djauhar I, Seswandhana R, et al. A

598

community based randomized controlled trial of iron and zinc supplementation in Indonesian

599

infants: effects on growth and development. Am J Clin Nutr, 2004; 80: 729-736.

600

[37] Nolunkcwe B. Child Nutritional Status and Household Patterns in South Africa 2007;

601

volume 7, no 5.

602

[38] National Food Consumption Survey- Fortification Baseline. The knowledge, attitude,

603

behaviour and procurement regarding fortified foods, a measure of hunger and the

604

anthropometric and selected micronutrient status of children aged 1-9 years and women of child

605

bearing age; South Africa, Pretoria; Directorate; Nutrition, Department of Health; 2005.

606

[39] Uchida T, Miyanaga Y, Tanaka H, Wada K, Kurosaki S, Ohki T, et al. Quantitative

607

evaluation of the bitterness of commercial medicines using a taste sensor. Chem Pharm Bull,

608

2000; 48 (11): 1843-1845.

609

[40] Tanaka M. Secretory function of the salivary gland in patients with taste disorders or

610

xerostomia: Correlation with zinc deficiency. Acta Otolaryngol, 2002; Suppl 546: 134–141.

611

[41] Manary MJ, Hotz C, Krebs NF, Gibson RS, Westcott JE, Broadhead RL, et al. Zinc

612

homeostasis in Malawian children consuming a high-phytate, maize-based diet. Am J Clin Nutr,

613

2002; 75: 1057-1061.

614

[42] Manary MJ, Hotz C, Krebs NF, Gibson RS, Westcott JE, Arnold T, et al. Dietary phytate

615

reduction improves zinc absorption in Malawian children recovering from tuberculosis but not in

616

well children. J Nutr, 2000; 130: 2959-2964.

617

[43] Brown KH, Peeraon JM, Rivera J, Allen LH. Effect of supplemental zinc on the growth and

618

serum zinc concentrations of prepubertal children: a meta-analysis of randomized controlled

619

trials. Am J Clin Nutr, 2002; 75 (6): 1062-1071.

620

[44] Maret W, Sandstead HH. Zinc requirements and the risk and benefits of zinc

621

supplementation. J Trace Elem Med Biol, 2006; 20:3-18.

622

[45] UNICEF. Progress report on children’s nutrition No 4; 2006.

623

[46] Sreedhar B, Subramaniyan R, Nair KM. A protective role for zinc on intestinal peroxidative

624

damage during oral iron repletion. Biochem Biophys Res Commun, 2004; 318: 992-997.

625

[47] Walker CR, Kordas K, Stolzfus RJ, Black RE. Interactive effects of iron and zinc on

626

biochemical and functional outcomes in supplementation trials. Am J Clin Nutr, 2005; 82: 5-12.

AC C

EP

TE D

M AN U

SC

RI PT

597

21

ACCEPTED MANUSCRIPT

[48] Sandstrom B, Davidsson L, Cederblad A, Lonnerdal B. Oral iron, dietary ligands and zinc

628

absorption. J Nutr, 1985; 115: 411–414.

629

[49] Oelofse A, Van Raaij JMA, Benadé AJS, Dhansay MA, Tolboom JJM, Hautvast JGAJ.

630

'Disadvantaged black and coloured infants in two urban communities in the Western Cape, South

631

Africa differ in micronutrient status', Public Health Nutr, 2002; 5(2): 290-294.

632

[50] Tatala S, Svanberg U, Mduma B. Low dietary iron availability is a major cause of anemia: a

633

nutrition survey in the Lindi District of Tanzania. Am J Clin Nutr, 1998; 68: 171-178.

634

[51] Nel JH, Steyn NP. Report on South African Food Consumption Studies Undertaken

635

Amongst Different Population Groups (1983-2000): Average Intakes of Foods Most Commonly

636

Consumed. Pretoria: Department of Health, 2002.

AC C

EP

TE D

M AN U

SC

RI PT

627

22

ACCEPTED MANUSCRIPT

Table 1: Median and Interquartile Range of anthropometric and biochemical measurements of stunted and non-stunted children (n= 349). Total (n=349) Stunted (n=65)

Non-Stunted (n=284)

P-Values

Age (years)

4.22 (0.66)

4.61 (1.08)

4.56 (0.91)

0.552

Weight (kg)

16.00 (2.50)

16.00 (2.50)

17.00 (2.00)

<0.0001

Height (cm)

99.95 (7.45)

94.20 (8.03)

103.52 (7.48)

<0.0001

WHZ

0.56 (6.48)

0.29 (2.69)

0.13 (1.37)

0.358

HAZ

-1.02 (1.33)

-2.64 (0.89)

-0.77 (1.05)

-

WAZ

-0.23 (4.83)

-1.01(1.57)

-0.01(0.96)

<0.0001

BAZ Hb (g/dl)

0.58 (1.40) 11.40 (1.1)

1.09 (3.11) 11.50 (1.00)

0.32 (1.32) 11.40 (1.1)

<0.0001 0.615

Iron (µmol/l) Transferrin (g/l)

13.20 (8.3) 2.60 (0.5)

13.40 (6.20) 2.50 (0.4)

13.90 (9.3) 2.60 (0.4)

0.199 0.207

Transferrin saturation (%) Ferritin (ng/ml)

21.00 (13.0) 25.00 (18)

20.00 (11.5) 29.00 (16.50)

21.00 (14.0) 26.00 (17.50)

0.272 0.632

CRP (mg/l) Zinc (µmol/l)

0 (2.00) 10.15 (4.5)

0 (3.00) 11.40 (3.30)

0 (2.00) 10.30 (5.10)

0.137 0.071

M AN U

SC

RI PT

Variables

AC C

EP

TE D

WAZ: Weight for age; HAZ: Height for age; WHZ: Weight for height; BAZ: BMI for age

ACCEPTED MANUSCRIPT

Table 2: Nutritional status of the children as defined by their anthropometric indices Total (n=349) n %

Girls (n=186) n %

Boys (n=163) n %

p-value

Wasting

5

1.4%

0

0

5

3.1%

Χ2=5.788; p=0.016

Stunting

65

18.6%

30

16.1%

35

21.5%

Χ2=1.636; p=0.200

Underweight

1

0.3%

0

0

1

0.6%

Χ2=1.163; p=0.286

Overweight

73

20.9%

39

21%

34

20.9%

Χ2=0.0006; p=0.980

Obese

34

9.7%

18

9.7%

16

9.8%

Χ2=0.002; p=0.965

AC C

EP

TE D

M AN U

SC

RI PT

Anthropometric Indicator

ACCEPTED MANUSCRIPT

Table 3: Prevalence of micronutrient deficiencies as determined by the selected biochemical variables Cut-off points

Total (n=349)

Girls (n=186)

Boys (n=163)

p-value

Deficient

Normal

Deficient

Normal

Deficient

Normal

43 (26.4%)

120 (73.6%)

11 (6.7%)

152 (93.3%)

45 (27.6%)

118 (72.4%)

65 (39.9%)

98 (60.1%)

<11g/dl

98 (28%)

251 (72%)

55 (29.6%)

131(70.4%)

Ferritin

<12µg/dl

25 (7.2%)

324 (92.8%)

14 (7.5%)

172 (92.5%)

Transferrin saturation

<15%

98 (28.1%)

251 (71.9%)

53 (28.5%)

133 (71.5%)

Zinc

<9.9µg/dl

149 (42.6%)

200 (57.3%)

84 (45.2%)

102 (54.8%)

AC C

EP

TE D

M AN U

SC

Haemoglobin

RI PT

Variables

Χ2=0.437; p=0.508 Χ2=0.079; p=0.778 Χ2=0.034; p=0.854 Χ2=0.991; p=0.319

ACCEPTED MANUSCRIPT

Table 4: Zinc Taste Test results of the children by gender Girls (n=186) N (%) 8 (4.3%) 60 (32.3%) 118 (63.4%) 1 (0.5)

Boys(n=163) N (%) 6 (3.7%) 45 (27.6%) 112 (68.7%) 1 (0.5)

AC C

EP

TE D

M AN U

SC

Taste bad Little taste than sweet Tasteless Symptoms

Total (n=349) N (%) 14 (4%) 105 (30%) 230 (65.9%) 2 (0.5)

RI PT

Description of taste

ACCEPTED MANUSCRIPT

Table 5: Correlation between biochemical markers and anthropometric measurements WHZ

HAZ

WAZ

BAZ

CRP

Zinc

r

-0.071

0.015

-0.043

-0.051

p

0.185

0.774

0.514

0.341

r

0.169

0.045

0.136

0.124

p

0.002

0.398

0.038

0.021

r

-0.094

0.025

-0.047

p

0.079

0.648

0.477

r

-0.104

-0.018

-0.070

-0.056

0.128

0.079

p

0.052

0.738

0.289

0.297

0.017

0.139

r

-0.025

-0.145

-0.114

-0.017

-

-0.112

p

0.638

0.007

0.084

0.753

r

-0.174

-0.122

-0.116

-0.104

-0.112

p

0.001

0.052

0.037

Ferritin

Zinc

TE D

CRP

0.022

0.079

-0.110

-0.093

0.040

0.083

-0.061

-0.277

0.257

<0.0001 0.045

M AN U

Tsaturation

EP

0.107

0.037

WAZ: Weight for age; HAZ: Height for age; WHZ: Weight for age; BAZ: BMI for age

AC C

0.103

<0.0001 0.055

SC

Transferrin

-0.304

RI PT

Iron

-

ACCEPTED MANUSCRIPT

Table 6: Correlations between biochemical markers and anthropometric measurements in non-stunted children HAZ

WAZ

-0.051

-0.039

-0.136

p

0.390

0.512

r

0.175

p

CRP

Zinc

-0.052

-0.288

0.130

0.063

0.380 <0.0001

0.028

-0.003

0.156

0.165

-0.133

-0.075

0.003

0.957

0.033

0.005

0.025

0.208

r

-0.083

-0.020

-0.161

-0.082

-0.266

0.140

p

0.163

0.733

0.027

SC

r

BAZ

RI PT

WHZ

0.170 <0.0001

0.018

r

-0.148

0.007

-0.118

-0.119

0.115

0.085

p

0.013

0.903

0.107

0.045

0.054

0.152

r

-0.058

-0.137

-0.087

-0.065

-

-0.109

p

0.331

0.021

0.239

0.276

r

-0.216

-0.078

-0.145

-0.190

-0.109

<0.0001

0.191

0.048

<0.001

0.067

Iron

Transferrin

Ferritin

Zinc p

TE D

CRP

M AN U

Tsaturation

0.067

AC C

EP

WAZ: Weight for age; HAZ: Height for age; WHZ: Weight for age; BAZ: BMI for age

-

ACCEPTED MANUSCRIPT

1

Table 7: Linear regression models for assessing selected biochemical predictors of malnutrition states in children

0.024 0.051

<0.0001 0.032

95% CI

RI PT

p-value

-0.043, -0.003 -0.069, 0.000

-0.695, -0.335 -0.071, -0.003

0.035

-0.403, -0.015

M AN U

Variables Equation of regression Stunting as assessed by HAZ CRP (x) -0.699-0.122 (x) Zinc (x) -0.699- 0.105 (x) 2 R=0.150, R =0.023, SEE=1.181, p=0.019 Wasting as assessed by WHZ Age (x) 2.247 -0.288 (x) Zinc (x) 2.247 – 0.110 (x) R=0.337, R2=0.113, SEE=1.167, p< 0.0001 Overweight as assessed by BAZ Age (x) 1.509 – 0.209 (x) 2 R=0.113, R =0.013, SEE=1.268, p=0.035

SC

2 3

4 5

AC C

EP

TE D

6

1

ACCEPTED MANUSCRIPT

Highlights

RI PT

SC

M AN U

•

TE D

•

EP

•

The study examines the prevalence of Iron and Zinc deficiencies in preschool children which are among the vulnerable groups. This presence of iron and zinc deficiencies was accompanied by both stunting and overweight and obesity. The study shows that these children might be experiencing problems of over and undernutrition which are accompanied by hidden hunger as indicated by the micronutrient deficiencies. The children also had anaemia which may impact on their cognitive development especially when accompanied by iron deficiency.

AC C

•