Food Chemistry 255 (2018) 365–371

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Heavy metal content and element analysis of infant formula and milk T
powder samples purchased on the Tanzanian market: International branded
versus black market products
⁎
M. Sagera, C.R. McCullochc, D. Schoderb,c,
a
Special Investigations in Element Analysis, Austrian Agency for Health and Food Safety, Spargelfeldstrasse 191, 1220 Vienna, Austria
b
Institute of Milk Hygiene, University of Veterinary Medicine, Veterinärplatz 1, 1210 Vienna, Austria
c
Veterinarians Without Borders, Austria

A R T I C L E I N F O A B S T R A C T

Keywords: Milk powder is a food for malnourished African children and for healthy infants of women with HIV/AIDS. High
Milk powder demand and low purchasing power has resulted in a huge informal, black market in Sub-Saharan Africa. Forty-
Element analysis three milk powder batches were analyzed for 43 chemical elements using ICP-MS One sample (2.3%) was
Heavy metal contaminated at a lead concentration of 240 µg/kg dry weight exceeding the European threshold (130 µg/kg dry
Lead
weight). Macroelement contents revealed a trend decreasing in concentration through skimmed, full cream
Iron
products to infant formulae. Concentration ranges by dry weight differed in respect of uncertainty intervals of
Baby formula
Tanzania ± 10%. Median Ca, K and P concentrations declined from 11.14 g/kg to 3.21 g/kg, 14.11 g/kg to 4.95 g/kg and
9.12 g/kg to 2.75 g/kg dry mass, respectively. Milk powder samples obtained from the Tanzanian black market
were comparable in respect of nutritional and chemical content to international branded full cream products.

1. Introduction millions of malnourished Africans, it is also recommended as a breast

The macro- and micronutrient contents of milk are known to vary
according to several variables, such as breed of cow, lactation status,
season, herd health and nutrition (Benincasa, Lewis, Sindona, &
Tagarelli, 2008, Brescia, Caldarola, Buccolieri, & DelĺAtti & Sacco,
2003, Coni, Bocca, Ianni, & Caroli, 1995, Knowles, Grace, Knight,
McNabb, & Lee, 2006, Sager & Hobegger, 2013). Nutrition is probably
the most important variable because of the ease with which it can be
modified and is dependent on a range of local infrastructure and geo-
graphic variables, such as soil, season and climate.
Milk is also one of the main consumable products of animal origin
and it is particularly important for survival in overpopulated regions of
the world, such as sub-Saharan Africa, where meat or other sources of
protein are rarely available (Schoder, Maichin, Lema, & Laffa, 2013).
However, its suitability for consumption rapidly diminishes if cooling
cannot be provided. To circumvent this problem, and to make it easier
to transport, powdering can stabilize milk. However, there are only a
few milk powder production plants in sub-Saharan Africa and milk
scarcity is rectified by imported milk powder. Demand is such that up to
55% of the world’s milk powder production is sold to the African
continent each year (Food and Ariculture Organisation of the United
Nations, 2013). Imported milk powder is not only a food aid for
milk substitute for healthy infants of women with HIV/AIDS. Alar-
mingly, some 30–50% of all HIV infections in African infants result
from breast-feeding by infected mothers (WHO, 2010).
These factors are especially visible in Tanzania that is both highly
populated and well served at Dar es Salaam by international shipping.
According to World Bank statistics for 2014, gross national income per
capita per month in Tanzania was US$77 and the population percen-
tage living below the national poverty line, the so called poverty
headcount ratio, was 28.2% (World Bank, 2016). Therefore, a sub-
stantial number of families cannot afford to buy officially imported
powdered milk at Western prices, and are forced to purchase milk
powder on the informal black market (Schoder, 2010). However, due to
the international bulk buying and mixing practices of the milk powder
industry, it is very difficult to source the raw milk. This is compounded
by a poorly developed food surveillance system in sub-Saharan Africa,
which is unable to monitor food sources effectively and thereby safe-
guard public health. These short-comings are illustrated by previous
studies showing that Africa was largely defenseless against the Chinese
melamine scandal; six infants died and 294,000 children developed
renal stones as a consequence (Schoder 2010, 2012, 2016). According
to Schoder (2010), 11% of the milk powder samples purchased in
Tanzania contained melamine at dangerous levels suggesting

⁎
Corresponding author at: Institute of Milk Hygiene, University of Veterinary Medicine, Veterinärplatz 1, 1210 Vienna, Austria.
E-mail address: Dagmar.Schoder@vetmeduni.ac.at (D. Schoder).

https://doi.org/10.1016/j.foodchem.2018.02.058
Received 30 June 2017; Received in revised form 6 October 2017; Accepted 11 February 2018
Available online 16 February 2018
0308-8146/ © 2018 Elsevier Ltd. All rights reserved.
M. Sager et al.
contaminated milk from China had made its way to East Africa, due in
part to lack of efficient food surveillance systems in Africa and China.
Therefore, given all these deficiencies, this study sought to: (i) de-
termine the elemental composition and heavy metal contamination of
milk powder products purchased on the formal and informal Tanzanian
markets, (ii) check whether these products meet legal, health, and
nutritional requirements, and (iii) compare nutritional values with
Austrian reference samples.
2. Material and methods
2.1. Tanzanian milk powder supply and marketing systems
Tanzanian milk marketing systems can be categorized by the type of
milk powder distributed and by how it is sold. Milk powder types in this
study comprised: (i) branded products of international and multi-
national companies, and (ii) illegally imported, unlabeled products sold
openly or discretely on local black markets. Powders are sold through
retail and modern distribution outlets, such as: (i) supermarkets, which
sell exclusively imported and authorized branded products, and (ii)
street vendors and micro-retailers promoting illegally imported, un-
labeled black market products. Supermarket customers in the sampled
regions are typically high-income earners (expatriates and nationals),
whereas street vendors and micro-retailers sell to low-income con-
sumers at half the usual retail price or less.
2.2. Sampling and types of milk powder purchased on the Tanzanian
market
Two categories of milk powder samples were selected: (i) branded
products of international companies from authorized channels
(n = 20), and (ii) illegally imported, unlabeled products (n = 23).
Branded products were purchased directly from a variety of super-
markets. These products were selected on the following basis: (i) im-
mediate availability, (ii) product category (milk powder-infant for-
mula), (iii) type of manufacturer, and (iv) production and expiry date.
Branded milk powder products could be traced to eleven international
and multinational companies supplying the East African market
(Table 1): (i) Pharmacare, Ltd., Sandton, South Africa; (ii) Promasidor,
Ltd., Zug, Switzerland; (iii) Nestle (South Africa) (Pty.), Ltd., Randburg,
South Africa; (iv) Daily Fresh Foods, Ltd., UK; (v) Irish Dairy Board,
Dublin, Republic of Ireland; (vi) Clover S.A. (Pty.), Ltd., Roodepoort,
South Africa; (vii) Chivers Ireland, Ltd., Dublin, Republic of Ireland;
(viii) Cow & Gate Ireland, Dublin, Republic of Ireland; (ix) Nestle Foods
Kenya, Ltd., Nairobi, Kenya; (x) National Food Industries Co., Ltd.,
Jeddah, Kingdom of Saudi Arabia; and (xi) Wyeth Nutritionals, As-
keaton, Co., Limerick, Ireland.
Unlabeled products (n = 23) were purchased from micro-retailers
on the local black market (Schoder, 2010, 2012, 2016). According to a
recent study, 300 kiosks and micro-retailers sell milk powder to low-
income consumers in Dar-es-Salaam (Schoder, 2016). The originally
packed samples were sent to the University of Veterinary Medicine,
Vienna, Austria, by air cargo, before dispatch to the analytical labora-
tory at the Austrian Agency for Health and Food Safety (AGES).
2.3. Digestions and inductively-coupled-plasma (ICP) multi-element
analysis
Aliquots of the milk powder samples were transferred to screw-
capped polypropylene vials and transported to the analytical laboratory
at the AGES, where the multi-element analysis was performed. Since
boron and silicon were also determined, contact with glass was strictly
avoided. Samples were weighed into PTFE (polytetrafluoroethylene)-
lined pressure bombs, digestion reagents added and mixtures micro-
wave-heated.
Two independent digestion methods were used and performed in
366
Food Chemistry 255 (2018) 365–371
duplicate. These involved either nitric acid or potassium chlorate
(KClO3). While most of the elements can be determined by either
method, for cationic reasons KClO3 digestion is not appropriate for K
and Rb. KClO3 digests are especially suitable for B, S, Si and I analyses
and develop significantly lower internal pressures. Samples weighing
0.25 g were digested with 3.8 ml of 65% HNO3 plus 0.1 ml of 40% HF
and made up to 25 ml. In parallel, 1.00 g samples were digested with
8 ml of 7% KClO3 solution in 4.5 M HNO3 (20 g KClO3 p.a. + 200 ml
H2O + 80 ml suprapure HNO3), and also made up to 25 ml, according
to Sager (2011).
A MLS 1200 mega high performance microwave digestion unit (MLS
GmbH, D-88299 Leutkirch) was used and programmed as follows: three
minutes at 250 W/two minutes zero/five minutes 250 W/five minutes
400 W/five minutes 500 W. After cooling, the contents of the pressure
bombs were made up to 25 ml in polypropylene volumetric flasks. A
Perkin-Elmer Optima 3000 XL instrument with an axial torch was used
for ICP- multi-element determinations. Matrix-matched calibrants were
used for determinations following KClO3- digests. Non-metals and La-
Ce-Y-Sc-Ti were determined in separate runs. ICP-MS measurements for
Bi, Cd, Co, Mo, Ni, Pb, Tl, Y and the rare earth elements were performed
with a Perkin Elmer Sciex ELAN DRC II at 1 + 4 dilution after addition
of indium (1 mg/l sample) as internal standard. Total I was measured in
a separate run as iodate after a 1 + 49 dilution with 1/80 diluted 7%
KClO3 digestion solution (described above) and evaluated by standard
addition.
2.4. Legal requirements
The European Union has so far set thresholds for intake levels of the
potentially hazardous metals lead, cadmium and mercury (European
Commission, 2006). For many other elements, such as copper, zinc,
chromium, iron, manganese and cobalt, levels that lead to deficiencies
and chronic or acute poisonings are known, but no explicit thresholds
have yet been set.
However, the Nutrition Societies of Germany, Austria and
Switzerland have established various threshold values (see DACH
(2013) for further details).
2.5. Element concentration ratios
Due to weathering and geochemical transport of rocks, it is plau-
sible that fractionations might occur naturally between elements with
similar chemical properties (Wedepohl, 1995). To ascertain if this has
some bearing on the geographical origin of milk samples, concentration
ratios of elements were compared. Geochemically-relevant ratios are K/
Rb and Ca/Sr. In contrast, other ratios of interest, such as Fe/Mn, Ca/
Mg, P/S and Cu/Mo, which may have relevance to bovine health and
nutrition, were also examined.
2.6. Element comparison between Tanzanian milk powder samples and
Austrian reference samples
Elemental compositions of the Tanzanian milk powder samples
were compared directly with a set of 104 Austrian reference samples.
These were individual milking samples from Fleckvieh and Holstein
Friesian cows collected directly from the farm (Sager & Hobegger,
2013). The farms were situated within three geologically diverse re-
gions: (i) granite-gneiss; (ii) tertiary-quaternary sediments, and (iii)
flysch and limestone. The samples were collected in 120 ml screw-
capped, plastic vials, deep frozen and freeze-dried. They were digested
with concentrated nitric acid (0.25 g of sample + 3.8 ml of HNO3) and
potassium chlorate-dilute nitric acid (1.0 g of sample + 8 ml of diges-
tion solution) in pressure bombs by microwave-assisted heating, and
made up to 25 ml in plastic volumetric flasks.
 M. Sager et al.

Table 1
List of all tested branded milk powder and infant formula products of international and multinational companies supplying the East African market (n = 20).

N° Product Name Manufacturer Expiry date Production date Lot n° Skimmed milk Full fat cream Infant formula

a
1 Aspen™ – Nutritionals S-26/1 (Infant formula with iron) Pharmacare Ltd., Sandton, South Africa 2010–05 2008–05 502472 1
2a Nestle® Lactogen®/1 (Starter infant formula with iron) Nestle (South Africa) (Pty) Ltd., Randburg, South Africa 2009–12 2007–12 73400179L1 1
3c Wyeth® – S-26 Gold Stage 1 (Infant formula from birth to Wyeth Nutritionals, Askeaton, Co., Limerick, Ireland 2009–08 2008–02 8C295611 1
6 months)
4a Nestle® NAN®/1 (Starter infant formula with iron) Nestle (South Africa) (Pty) Ltd., Randburg, South Africa 2009–02 2007–11 7325017981 1
5a Nestle® NIDO® (Full cream milk powder) Nestle (South Africa) (Pty) Ltd., Randburg, South Africa 2010–02 2008–02 80560181 1
6a Nestle® NAN®/1 (Starter infant formula with iron) Nestle (South Africa) (Pty) Ltd., Randburg, South Africa 2010–03 2008–03 8065017981 1
7c Kerry Gold® (Full cream milk powder) Irish Dairy Board, Dublin, Republic of Ireland 2010–05 2007–05 02K/400 1
8b Cowbell® (Instant filled milk powder) Promasidor Ltd., Zug, Switzerland 2009–07 2008–04 5141A 1
9a Nestle® Lactogen®/2 (Follow-up formula with iron and Nestle (South Africa) (Pty) Ltd., Randburg, South Africa 2010–07 2008–07 82040181 SWWBFM 1
prebiotics)
10a Elite® (Fat free instant milk powder) Clover S.A. (Pty) Ltd., Roodepoort, South Africa 2009–04 NAe HO91-8-(3)-2 1
11a Aspen™ Nutritionals- S-26 Progress/3 (Growing up drink for Pharmacare Ltd., Sandton, South Africa 2010–07 2008–07 502535 1
children)

367
c e
12 Marvel (Original dried skimmed milk) Chivers Ireland Ltd., Dublin, Republic of Ireland 2009–03 NA 7268 1
13c Daily Fresh™ (Full cream milk powder) Daily Fresh Foods Ltd., U.K. 2009–12 2008–06 AM/IMP/14B 1
14c Cow & Gate®/2 (Nutritionally complete infant milk) Cow & Gate Ireland, Dublin, Republic of Ireland 2009–10 NAe 1251 1
15a Elite® (Fat free instant milk powder) Clover S.A. (Pty) Ltd., Roodepoort, South Africa 2009–03 NAe HO63-8-(3)-3 1
16b Cowbell® (Instant filled milk powder) Promasidor Ltd, Zug, Switzerland 2010–01 2008–09 4381B 1
17b Nestle® NIDO® (Full cream milk powder) Nestle Foods Kenya Ltd., Nairobi, Kenya 2009–03 2007–09 72700169 A1 1
18b Nestle® NIDO® (Full cream milk powder) Nestle Foods Kenya Ltd., Nairobi, Kenya 2009–03 2008–03 80720169 D1 1
19d Luna (Instant full cream milk powder) National Food Industries Co. Ltd., Jeddah, Kingdom of Saudi 2009–09 2008–04 104 NZMP DR 27 1
Arabia
20b Nestle® NIDO® (Full cream milk powder) Nestle Foods Kenya Ltd., Nairobi, Kenya 2009–03 2008–03 80780169 D1 1

3 9 8

a
Produced in Africa.
b
only packaged in Africa.
c
produced or packaged in Europe.
d
produced or packaged in New Zealand.
e
data not available (NA).
Food Chemistry 255 (2018) 365–371
M. Sager et al. Food Chemistry 255 (2018) 365–371

Table 2a
Main and essential trace element concentrations (mg/kg dry mass) of 43 milk powder batches collected in Dar-es-Salaam, Tanzania, East Africa.

Formal market (n = 20) Informal (African) market (n = 23)

Skimmed milk (n = 3) Full cream milk (n = 9) Infant formulas (n = 8)

Element Median Min – Max Median Min – Max Median Min – Max Median Min – Max

Ca 11,141 11,090–11,479 7666 6443–9152 3214 2820–6630 8477 7205–8867

Co 0.061 0.027–0.064 0.041 0.026–0.061 0.028 0.018–0.036 0.041 0.033–0.063
Cr 0.029 0.014–0.069 < 0.007 < 0.007–0.040 < 0.007 < 0.007–0.053 0.076 0.060–0.091
Cu 0.439 0.393–0.473 0.312 0.054–0.592 3.45 0.41–6.35 0.405 0.218–0.664
Fe 3.84 2.93–4.00 28.7 1.64–134 58.9 3.6–77.8 2.45 1.91–3.38
J 1.87 1.64–3.25 0.98 0.14–4.74 2.71 1.35–3.59 2.08 1.13–6.38
K 14,105 13,939–14,686 9545 6981–11,027 4951 3868–11,131 10,452 8988–11,386
Mg 1042 1021–1073 699 653–952 397 351–688 776 701–806
Mn 0.367 0.282–0.484 0.337 0.142–0.884 0.517 0.157–0.796 0.184 0.157–0.265
Mo 0.242 0.238–0.341 0.268 0.246–0.310 0.127 0.050–0.175 0.308 0.296–0.383
Na 3868 3717–3906 2630 2408–3731 1711 1076–2787 2777 2603–3419
P 9121 9012–9656 6055 5712–7018 2755 1888–5372 6843 5923–7411
S 3556 3542–3759 2328 861–2964 1483 1201–2024 2331 2021–2759
Zn 45.7 41.7–49.3 28.1 13.9–85.2 39.6 23.6–53.9 31.4 26.8–33.8

Table 2b
Concentration of toxic elements and elements of low physiological activity, i.e. rare earth elements (mg/kg dry mass) of 43 milk powder batches collected in Dar-es-Salaam, Tanzania,
East Africa.

Formal market (n = 20) Informal (African) market (n = 23)

Skimmed milk (n = 3) Full cream milk (n = 9) Infant formulas (n = 8)

Element Median Min – Max Median Min – Max Median Min – Max Median Min – Max

x
Ba 1.203 0.673–1.366 0.458 0.269–1.941 0.337 0.138–0.498 0.594 0.459–0.751
Cdx 0.003 < 0.001–0.003 0.002 < 0.001–0.004 0.002 < 0.001–0.007 0.004 0.002–0.008
Nix 0.08 0.04–0.13 0.03 < 0.02–0.15 0.05 < 0.02–0.17 < 0.02 < 0.02–0.05
Pbx 0.02 0.01–0.03 < 0.01 < 0.01–0.24 < 0.01 < 0.01–0.05 < 0.01 < 0.01–0.04
Al 2.2 1.6–2.7 1.85 0.5–3.8 1 0.6–2.3 < 0.5 < 0.5–1.1
B 3.53 0.83–4.12 0.78 0.27 –1.84 0.5 0.09–0.76 0.74 0.21–1.55
Be < 0.004 < 0.004 < 0.004 < 0.004 < 0.004 < 0.004–0.006 < 0.004 < 0.004
Bi < 0.04 < 0.04 < 0.04 < 0.04–0.06 < 0.04 < 0.04 < 0.04 < 0.04–0.05
Ce 0.009 0.007–0.011 < 0.002 < 0.002–0.011 0.004 < 0.002–0.010 0.002 < 0.002–0.006
Cs 0.03 0.022–0.047 0.131 0.013–0.310 0.007 0.003–0.017 0.019 0.013–0.029
Er < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001
Eu 0.0002 0.0002–0.0004 0.0002 0.0001–0.0004 0.0001 0.0001–0.0003 < 0.0001 < 0.0001–0.0002
Gd < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001–0.001 < 0.001 < 0.001
Ho < 0.0001 < 0.0001 < 0.0001 < 0.0001 0.0001 < 0.0001–0.0003 < 0.0001 < 0.0001
La 0.003 0.003–0.006 0.002 < 0.002–0.010 0.005 < 0.002– 0.014 < 0.002 < 0.002–0.004
Li 0.087 0.038–0.089 0.046 0.004–0.095 0.038 0.010–0.075 0.032 0.007–0.057
Lu < 0.0003 < 0.0003 < 0.0003 < 0.0003 < 0.0003 < 0.0003 < 0.0003 < 0.0003
Nd 0.002 0.001–0.002 0.0012 < 0.001–0.004 0.0012 < 0.001–0.005 < 0.001 < 0.001–0.001
Pr < 0.001 < 0.001 < 0.001 < 0.001–0.001 < 0.001 < 0.001–0.001 < 0.001 < 0.001
Rb 22.01 13,36–31,05 18.37 6,40–35,56 3.54 1,37–9,57 6.55 5,88–13,67
Sc < 0.03 < 0.01–0.01 < 0.03 < 0.03 < 0.03 < 0.03 < 0.03 < 0.03
Si 13.21 4.77–20.7 7.61 3.86–10.84 7.1 3.67–13.4 2.98 2.31–5.60
Sm < 0.002 < 0.002 < 0.002 < 0.002 < 0.002 < 0.002 < 0.002 < 0.002
Sr 3.09 2.56–4.52 2.79 1.93–4.37 1.56 0.82–2.95 2.69 2.28–3.03
Tb < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001–0.0001 < 0.0001 < 0.0001
Ti 0.19 < 0.01–0.30 0.22 0.02–0.26 0.02 < 0.01–0.13 0.11 0.10–0.15
Tl < 0.001 < 0.0002–0.0007 0.002 < 0.001– 0.006 < 0.001 < 0.001–0.001 < 0.001 < 0.001–0.001
V < 0.02 < 0.02 < 0.02 < 0.02–0.11 < 0.02 < 0.02 < 0.02 < 0.02
Y 0.002 0.001–0.002 0.002 < 0.001– 0.002 0.004 < 0.001–0.014 < 0.001 < 0.001

x toxic elements.

3. Results 3.1. Heavy metal contaminants

A total of 43 milk powder samples were collected in Dar-es-Salaam,
Tanzania, and comprehensively analyzed for 43 chemical elements.
Of the batches collected there were 20 branded products from the
formal market, which were manufactured and/or marketed by 11 in-
ternational and multinational companies. However, 20% of the branded
samples (4/20) were deficiently labeled and did not display production
dates (Table 1). Products from the local black market (n = 23) were
completely unlabeled and packaged in simple, transparent plastic bags.
EU regulations stipulated that cadmium intake should not exceed
7 µg/kg of body weight over the course of one week, which approx-
imates to an intake of 500 µg for an adult, assuming complete bioa-
vailability (European Commission, 2006). Cadmium is a toxic element
and was encountered in the investigated milk powder samples over a
range from < 1 to 7 µg/kg of dry mass. Therefore in these samples it
was very low and comparable to Austrian reference samples (Table 2a).
A specific threshold for lead in milk has been set at 20 µg/kg fresh

368
M. Sager et al.
weight, corresponding to 130 µg/kg dry weight. This was exceeded in
one sample, which was a full cream milk powder packaged in the UK
(branded sample number 13; Table 1) that showed a lead concentration
of 240 µg/kg dry weight (Table 2a). All other milk powder samples
revealed lead concentrations between < 10 µg and up to 50 µg/kg dry
weight. Most of the rare earth elements, such as Be, Bi, Sc, Ti and V,
were below detection limits (Table 2a).
3.2. Elemental composition of branded versus black market milk powder
samples
Branded milk powder samples of international companies for au-
thorized sale were grouped as follows: (i) skimmed milk (three sam-
ples), (ii) full cream milk (nine samples), and (iii) infant formulae (eight
samples).
Macroelement (Al, B, Ba, Ca, Co, K, Li, Mg, Na, P, Rb, S) content
revealed a trend decreasing in concentration in the order of skimmed,
full cream products to infant formulae (see Tables 2a and 2b). Con-
centration ranges by dry weight differed in respect of uncertainty in-
tervals of ± 10%. Median Ca concentrations declined from 11.14 g/kg
in skimmed milk samples, to 7.6 g/kg in (full cream samples, to 3.21 g/
kg for infant formulae. Similarly, K concentrations decreased from
14.11 g/kg to 9.55 g/kg to 4.95 g/kg for these respective samples. Mg
concentrations decreased from 1.04 g/kg to 0.70 g/kg to 0.39 g/kg. P
concentrations decreased from 9.12 g/kg to 6.05 g/kg to 2.75 g/kg dry
mass (see Table 2b).
The baby formulae contained more Cu, Fe and Mn, but less Mo, Si
and Sr than the branded full cream and skimmed milk powders. On the
other hand, milk powders sampled from the black market contained
more Cr, Cu and I, but less Fe, Li, Mn, Rb and Si than the branded
products of international companies. However, respective concentra-
tion ranges overlapped (see Tables 2a and 2b).
3.3. Element concentration ratios
Element concentration ratios are shown in Table 3. These data are
from samples sold on the African market and they closely resembles the
full cream products marketed by international companies. The differ-
ence is reflected in the median value for K/Rb, which was three times
higher (K/Rb = 1602 versus 492) in black market samples compared
with branded full cream milk product samples. Some labeled milk
products were Fe-fortified, which was not stated on the product label
(66.7%, 10/15, data not shown). This is revealed by the higher Fe/Mn
concentration ratios for infant formulae (114) compared with skimmed
milk powder (10.4), full cream powder (69.6) and milk powder from
the black market (13.08). However, the Ca/Mg concentration ratio was
found to be within a relatively narrow range throughout the samples
investigated (7.0 < Ca/Mg < 15.0) and the Ca/Sr ratios were scat-
tered within a broader range (1367 < Ca/Sr < 6448).
Table 3
Selected weight-based element concentration proportions of 43 milk powder batches.
Formal market (n = 20)
Skimmed milk (n = 3) Full cream milk (n = 9)
Median Min – Max Median
K/Rb 633 454–1099 492
Ca/Sr 3716 2455–4351 2584
Fe/Mn 10.4 7.94–10.92 69.6
Ca/Mg 10.7 10.69–10.86 10.55
P/S 2.56 2.54–2.57 2.73
Cu/Mo 1.81 1.15–1.99 0.84
Food Chemistry 255 (2018) 365–371
3.4. Elemental composition of branded milk powder samples compared with
Austrian reference samples
The elemental compositions of the branded milk powder samples
were compared with Austrian reference samples (Tables 4a and 4b).
Branded full cream milk powder samples had lower concentrations of
Ca, P, Cr and Mo and higher concentrations of Fe and Sr. All other
element concentrations were either close to or below detection limits,
or were found to occur within the same ranges. In comparison, infant
formulae contained half the concentrations of K, Mg, Mo, Na, and S, and
a third less Ca than the Austrian reference samples. However, con-
centrations of Cu and Fe were 10–30 times higher in baby formulae.
4. Discussion
Since cows’ milk and infant formulae are frequently the main or
only source of food for infants and small children, it is essential that the
quality and nutritional status of these foods meet health requirements.
Young children have unique nutritional requirements for growth and
their digestive systems are especially efficient at absorbing essential
elements compared with adults (WHO, 2014). Consequently, knowl-
edge of reliable reference values for these elements in milk is important
if supplementary infant nutrition is to prove adequate.
The market for milk and infant formulae is substantial. The global
figure for 2014 was US$ 49.2 billion (InkedIn, 2014), and massive in-
ternational promotion of these products is succeeding in raising de-
mand. This is particularly so in many low-income countries, including
Tanzania, where annual sales growth is estimated at around 10%
(Trading Economics, 2016). This contrasts with high-income countries
where growth is all but stagnant. Nevertheless, increased feeding of
these products to young children is not without concern (Piwoz, 2015).
Reports have surfaced about milk powder and infant formulae being
contaminated with toxic metals and environmental impurities
(Schümann, 1990). The relevance of cows’ milk for infants is also cri-
ticized, as discussed below.
Due to their low body-weights in the face of relatively high nutrient
requirements, infants are particularly sensitive to ingested con-
taminants. Efficient gastrointestinal absorption prior to full develop-
ment of internal organs, particularly liver and kidneys, can mean that
the effects of contaminated feed last a life-time (Oskarsson, Palminger
Hallén, Sundberg, & Petersson Grawé, 1998). Yet, it is not uncommon
for infants to be fed the same product brand daily for weeks or even
months at a time.
In an effort to safeguard health, the European Union has set
threshold intake levels of heavy metals, including lead, cadmium and
mercury (European Commission, 2006). In this respect, the threshold
for lead in milk is 20 µg/kg fresh weight, corresponding to 130 µg/kg
dry weight. In our study this value was exceeded in one sample, which
was actually a legitimate full cream milk powder that was identified as
being packaged in the UK and had a lead concentration of 240 µg/kg
dry weight. Concentrations of other heavy metals were unremarkable.
Informal (African) market (n = 23)
Infant formulas (n = 8)
Min – Max Median Min – Max Median Min – Max
308–1632 1365 849–3398 1602 652–3306
1858–3513 2235 1625–6448 3147 1442–3655
8.29–330 114 15.0–289 13.08 9.63–38.9
8.83–11.70 8.59 7.00–15.00 11 9.82–11.80
2.37–6.80 1.93 1.47–2.65 2.84 2.57–3.45
0.19–2.28 41.1 16.0–77.6 1.32 0.42–2.18
369
M. Sager et al. Food Chemistry 255 (2018) 365–371

Table 4a
Concentration Median and concentration ranges of main and essential trace elements (mg/kg dry mass) from Austrian reference samples and branded milk powder samples collected sold
in Dar-es-Salaam, Tanzania.

Lower Austria (n = 104) Full fat cream (n = 9) Infant formulas (n = 8)

Median Range Median Range Median Range

Ca 10,491 8434–13,679 7666 6443–9152 3182 2820–5659

Co 0.034 0.018–0.046 0.041 0.026–0.061 0.028 0.018–0.036
Cr 0.172 0.025–0.484 < 0.007 < 0.007–0.040 < 0.007 < 0.007
Cu 0.367 0.149–0.913 0.312 0.054–0.592 3.66 2.64–6.35
Fe 1.54 0.63–2.24 28.7 1.64–134 61.2 36.3– 77.8
J 1.78 0.65–4.77 0.98 0.14–4.74 3.06 1.35–3.59
K 10,492 8317–15,657 9545 6981–11,027 4893 3868–6944
Mg 791 646–964 699 653–952 386 351–501
Mn 0.206 0.085–0.374 0.337 0.142–0.884 0.529 0.269–0.796
Mo 0.352 0.244–0.857 0.268 0.246–0.310 0.127 0.050–0.175
Na 2724 1914–3771 2630 2408–3731 1505 1076–1895
P 7326 6178–9456 6055 5712–7018 2593 1888–4307
S 2561 2156–3498 2328 861–2964 1440 1201–1648
Zn 32.4 22.2–46.6 28.1 13.9–85.2 40.5 32.9–53.9

Table 4b
Concentration of toxic elements and elements of low physiological activity (mg/kg dry mass) from Austrian reference samples and branded milk powder samples collected in Dar-es-
Salaam, Tanzania.

Lower Austria (n = 104) Full fat cream (n = 9) Infant formulas (n = 8)

Median Range Median Range Median Range

x
Ba 0.52 0.26–0.95 0.458 0.269–1.941 0.318 0.138–0.495
Cdx 0.003 < 0.001–0.028 0.002 < 0.001– 0.004 0.002 < 0.001–0.007
Nix 0.03 < 0.02–0.185 0.03 < 0.02–0.15 0.05 < 0.02–0.17
Pbx < 0.01 < 0.01–0.07 < 0.01 < 0.01–0.24 < 0.01 < 0.01–0.05
Al < 0.5 < 0.5–1.49 1.85 0.5–3.8 1 0.6–2.3
B 0.81 0.231–1.94 0.78 0.27 –1.84 0.41 0.09–0.75
Be < 0.004 < 0.004 < 0.004 < 0.004 < 0.004 < 0.004–0.006
Bi < 0.04 < 0.04 < 0.04 < 0.04–0.06 < 0.04 < 0.04
Ce 0.003 < 0.002–0.064 < 0.002 < 0.002–0.011 0.005 < 0.002–0.010
Cs 0.013 0.003–0.084 0.131 0.013–0.310 0.006 0.003–0.017
Er < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001
Eu 0.0001 < 0.0001–0.0012 0.0002 0.0001–0.0004 0.0001 0.0001–0.0002
Gd < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001–0.001
Ho < 0.0001 < 0.0001–0.0004 < 0.0001 < 0.0001 0.0001 < 0.0001–0.0003
La < 0.002 < 0.002–0.019 0.002 < 0.002–0.010 0.004 < 0.002–0.009
Li 0.02 0.003–0.061 0.046 0.004–0.095 0.012 < 0.005–0.075
Lu < 0.0003 < 0.0003 < 0.0003 < 0.0003 < 0.0003 < 0.0003
Nd < 0.001 < 0.001–0.023 0.0012 < 0.001–0.004 0.002 < 0.001–0.005
Pr < 0.001 < 0.001–0.008 < 0.001 < 0.001–0.001 < 0.001 < 0.001–0.001
Rb 11.1 4.6–28.6 18.37 6.40–35.56 3.28 1.37–6.40
Sc < 0.03 < 0.03–0.04 < 0.03 < 0.03 < 0.03 < 0.03
Si 7.97 0.22–20.7 7.61 3.86–10.84 6.43 3.67–13.4
Sm < 0.002 < 0.002–0.006 < 0.002 < 0.002 < 0.002 < 0.002
Sr 1.95 1.01–2.86 2.79 1.93–4.37 1.41 0.82–2.95
Tb < 0.0001 < 0.0001–0.0006 < 0.0001 < 0.0001 < 0.0001 < 0.0001–0.0001
Ti 0.115 0.04–0.176 0.22 0.02–0.26 < 0.01 < 0.01–0.13
Tl < 0.001 < 0.001–0.003 0.002 < 0.001– 0.006 < 0.001 < 0.001–0.001
V < 0.02 < 0.02–0.03 < 0.02 < 0.02–0.11 < 0.02 < 0.02
Y < 0.001 < 0.001–0.014 0.002 < 0.001– 0.002 0.005 < 0.001–0.014

x toxic elements.

Lead is a potent neurotoxin and its absorption is increased when co-
administered with milk. Exposure to the fetus and infants irreversibly
affects development of the nervous system, resulting in reduced in-
telligence, learning disabilities (Dabeka, Fouquet, Belisle, & Turcotte,
2011) and anemia (Schümann, 1990). In comparison, studies elsewhere
have reported that typical lead concentrations in all types of infant
formulae range from 1 to 10 ng/g when reconstituted (Dabeka, 1989);
although levels as high as 143 ng/g (Hafez & Kishk, 2008) and even
450 ng/g (Winiarska-Mieczan & Kiczorowska, 2012) have been re-
ported. Moreover, instances of higher blood lead levels have been ob-
served in formula-fed infants compared with those that are breast-fed,
which may reflect contamination from storage cans or tainted tap water
(Rabinowitz, Leviton, & Needleman, 1985).
Contrary to our prior concerns about hygiene at sampling sites and
general milk quality in Tanzania (Schoder et al., 2013), and with the
exception of the aforementioned contaminated sample, legitimately
marketed, branded milk products of international companies and il-
legally imported, unlabeled products from unauthorized channels de-
monstrated similar inorganic compositions and qualities. The elemental
composition of the cheapest milk powders obtained from the Tanzanian
black market closely resembled those of the most expensive branded
full cream samples.
We identified several regularities on comparing formally marketed
and black market milk samples. Black market samples were unlabeled

370
M. Sager et al.
and packaged simply in transparent plastic bags. Nevertheless,
branding did not guarantee safety as it was a branded sample that
contained excessive lead. There was a tendency for black market sam-
ples to contain less of the essential trace elements Fe, Mn and Si com-
pared with branded products, albeit more Cr, Cu and I. Overlapping
concentration ranges nonetheless retract from any concern regarding
nutritional significance. Notwithstanding, the salient factor that dif-
ferentiates both product types is cost. International branded products
significantly challenge most African household budgets.
This study also set out to compare the elemental nutritional ade-
quacy of milk powders sampled in Tanzania and Austria. Results re-
vealed that in respect of element composition, with the exceptions of
those discussed below, samples were indeed similar. As requirements of
growing infants differ, this has prompted, for example, DACH joint
recommendations for Germany, Austria and Switzerland to establish
reference values for babies between 0 and 4 months, 5–12 months and
children 1–4 years of age (DACH, 2013). Assuming a daily energy re-
quirement of 550 kcal for a baby between 0 and 4 months (100 g of milk
dry mass; (DACH, 2013), recommended daily intakes are 220 mg Ca,
120 mg P, 24 mg Mg,100 mg Na, 0.5 mg Fe, 1 mg Zn and 0.2 mg Cu
(DACH, 2013). With regard to the macroelements Ca, P, Mg, Na and Zn,
all tested samples fulfilled DACH recommendations. Fe und Cu were
only specified for infant formula. However, our results reveal that from
the inorganic composition of Tanzanian labeled baby milk formulae,
milk was not the only ingredient. These formulae contained half as
much K, Mg, Mo, Na and S and a third less Ca than the Austrian re-
ference samples. This suggests additions of other components. Most
likely these additions are of plant origin, as green plants generally
contain more Mn and Mo and cereals contain more Mn than products of
animal origin and less Ca and S (Sager, 2010). Further, compared with
the Austrian reference samples and the branded full cream samples, the
infant formulae contained between 10 and 30-times higher Cu and Fe
concentrations.
These latter observations are relevant as numerous other studies
have shown that cows’ milk is deficient in iron with respect of the needs
of growing infants and young children (Ziegler, 2011). Cows’ milk
alone is associated with diminished iron stores and an increased risk of
iron deficiency (Ziegler, 2011). Although the iron content of cows’ milk
is comparable to that of human milk (0.5 mg/L), the large quantities of
casein and calcium in the former relative to human milk seriously im-
pair iron absorption. Cows’ milk has also been shown to induce in-
testinal occult blood loss in about 40% of otherwise healthy infants,
contributing to the risk of iron deficiency (Wilson, Lahey, & Heiner,
1974; Ziegler, 2011). Thus, when powdered cows’ milk products are
inadequately labeled, as was the case for unofficial sources in this
study, they are totally unsuitable for babies and infants. Of the Tan-
zanian samples analyzed in this study, 22.7% were defectively labeled.
Thus, the combination of factors that have given rise to the black
market for milk products in Tanzania and its consequences, specifically
missing or incomplete labeling, together with increased demand for a
breast milk substitute because of the risk of HIV transmission, are also
risk factors for infant iron deficiency.
5. Conclusion
According to the findings of this study, the elemental composition of
milk powder samples purchased from formal and informal markets in
Tanzania were essentially similar. Price was no guide to quality, as one
of the legitimate branded products contained excess lead. Comparison
with Austrian reference samples suggested that some samples were
fortified with iron and copper. Finally, the relevance of cows’ milk for
babies and infants is again called into question, particularly in respect
of heavy promotion of milk formulae in low-income countries, the de-
mand for a breast milk substitute on account of HIV/AIDS and the in-
adequacy of product labeling.
371
Food Chemistry 255 (2018) 365–371
References
Benincasa, C., Lewis, J., Sindona, G., & Tagarelli, A. (2008). The use of (32) multi element
profiling to differentiate between cow and buffalo milk. Food Chemistry, 110,
257–262.
Brescia, M. A., Caldarola, V., Buccolieri, G., Dell’Atti, A., & Sacco, A. (2003).
Chemometric determination of the geographical origin of cow milk using ICP–OES
data and isotopic ratios: A preliminary study. Italian Journal of Food Science, 15(3),
329–336.
Coni, E., Bocca, A., Ianni, D., & Caroli, S. (1995). Preliminary evaluation of the factors
influencing the trace element content of milk and dairy products. Food Chemistry, 52,
123–130.
Dabeka, R., Fouquet, A., Belisle, S., & Turcotte, S. (2011). Lead, cadmium and aluminum
in Canadian infant formulae, oral electrolytes and glucose solutions. Food Additives
and Contaminants Part A: Chemistry, Analysis, Control, Exposure and Risk Assessment,
28(6), 744–753.
Dabeka, R. W. (1989). Survey of lead, cadmium, cobalt and nickel in infant formulas and
evaporated milks and estimation of dietary intakes of the elements by infants 0-12
months old. Science of the Total Environment, 89(3), 279–289.
DACH. (2013). Reference values for nutrition supply. first ed. 5th reprint, DGE Bonn
2013. available at: https://www.dge.de/wissenschaft/referenzwerte/calcium/
Accessed 27.04.2016.
European Commission (2006). Commission regulation (EC) No 1881/2006 of Dec 19th
2006 on setting maximum levels for certain contaminants in foodstuffs. Official
Journal of the European Union, L364, 5–24.
Food and Ariculture Organisation of the United Nations. (2013). Market summaries: Milk
and milk products available at: http://www.fao.org/fileadmin/templates/est/
COMM_MARKETS_MONITORING/Dairy/Documents/November_2013.pdf 2013
Accessed 22.04.2016.
Hafez, L. M.1., & Kishk, A. M. (2008). Level of lead and cadmium in infant formulae.
Journal of Egypt Public Health Association, 83(3–4), 285–293.
Knowles, S. O., Grace, N. D., Knight, T. W., McNabb, W. C., & Lee, J. (2006). Reasons and
means for manipulating the micronutrient composition of milk from grazing dairy
cattle. Animal Feed Science and Technology, 131, 154–167.
InkedIn. (2014). Global Baby Food Market Insights, Trends & Analysis 2015–2022
available at https://www.linkedin.com/pulse/global-baby-food-market-insights-
trends-analysis-2015-tracy-black 2014 Accessed 27.04.2016.
Oskarsson, A., Palminger Hallén, I., Sundberg, J., & Petersson Grawé, K. (1998). Risk
assessment in relation to neonatal metal exposure. Analyst, 123(1), 19–23.
Piwoz, E. G. & Huffman, S. L. (2015). The impact of marketing of breast-milk substitutes
on WHO-recommended breastfeeding practices. 36(4), 373–86.
Rabinowitz, M., Leviton, A., & Needleman, H. (1985). Lead in milk and infant blood: A
dose-response model. Archives of Environment Health, 40(5), 283–286.
Sager, M. (2010). Über die elementzusammensetzung von in österreich erhältlichen fer-
tiggerichten und wurstwaren. Ernährung/Nutrition, 34(2), 57–64.
Sager, M. (2011). Microwave – assisted digestion of organic materials with KClO3/HNO3
for the analysis of trace metals and non– metals. Analytical Chemistry – An Indian
Journal, 10(2).
Sager, M., & Hobegger, M. (2013). Contents of elements in raw milk from three regions in
Lower Austria. Ernährung/Nutrition, 37, 277–290.
Schoder, D. (2010). Melamine milk powder and infant formula sold in East Africa. Journal
of Food Protection, 73(9), 1709–1714.
Schoder, D. (2012). Food adulteration with melamine on an international scale: Field
work and troubleshooting in Africa. In J. Hoorfar (Ed.). Case studies in food safety and
authenticity (pp. 308–317). Oxford: Woodhead ISBN: 978-0-85709-412-4.
Schoder, D., Maichin, A., Lema, B., & Laffa, J. (2013). Microbiological quality of milk in
Tanzania: From Maasai stable to African consumer table. Journal of Food Protection,
76(11), 1908–1915.
Schoder, D. (2016). Food fraud with melamine and global implications: Was Africa left
defenseless to the chinese melamine scandal? In J. Mei Soon (Pub.) (Ed.). Foodborne
Diseases. Case Studies of Outbreaks in the Agri-Food Industries (pp. 341–364). Boca
Raton: CRC Press ISBN: 978-1-4822-0827-6.
Schümann, K. (1990). The toxicological estimation of the heavy metal content (Cd, Hg,
Pb) in food for infants and small children. Z Ernahrungswiss, 29(1), 54–73.
Trading Economics Tanzania GDP Annual Growth Rate 2002–2016. available at http://
www.tradingeconomics.com/tanzania/gdp-growth-annual 2016 Accessed 27.04.
2016.
Wedepohl, K. H. (1995). The composition of the continental crust. Geochimica
Cosmochimica Acta, 59(7), 1217–1232.
WHO. (2010). Guidelines on HIV and infant feeding 2010. Principles and recommenda-
tions for infant feeding in the context of HIV and a summary of evidence. available at
http://www.who.int/maternal_child_adolescent/documents/9789241599535/en/
Accessed 29.06.2017.
WHO. (2014) Comprehensive implementation plan on maternal, infant and young child
nutrition. available at: http://www.who.int/nutrition/publications/CIP_document/
en/) Accessed 27.04.2016.
Wilson, J. F., Lahey, M. E., & Heiner, D. C. (1974). Studies on iron metabolism. V. Further
observations on cow's milk-induced gastrointestinal bleeding in infants with iron-
deficiency anemia. Journal of Pediatrics, 84(3), 335–344.
Winiarska-Mieczan, A., & Kiczorowska, B. (2012). Determining the content of lead and
cadmium in infant food from the Polish market. International Journal of Food Science
and Nutrition, 63(6), 708–712.
World Bank. (2016). World Development Indicators. available at: http://data.worldbank.
org/country/tanzania Accessed 21.04.2016.
Ziegler, E. E. (2011). Consumption of cow's milk as a cause of iron deficiency in infants
and toddlers. Nutrition Review, 69(1), 37–42.