Nutritional Quality, Level of Chemical Contaminants and Adulterants in Milk and Dairy Products in Ethiopia: A Review

Milk is a highly nutritious food, and it is a source of necessary macro and micro-nutrients for the growth, development and maintenance of human health. However the quality and safety of milk and dairy products become major health concern for consumers particularly to infants and children. This paper attempts to review the state of knowledge on nutritional quality, chemical contaminants and adulterants of milk and dairy products in Ethiopia. It also focuses on the method for analysis and identifies gaps for future work. Some work has been done in nutritional quality of milk and dairy products. The nutritional quality of Ethiopian milk products are substandard due to poor hygienic practices and water adulteration at farm level and value chain actors. Few of the research results reviewed for aflatoxin M1, organochlorine pesticide, heavy metals and antibiotic residues exceed international maximum permissible limit standards. The use of adulterants for economic gain is adding of water (dilution) in raw milk. In general there are limited reports in chemical contaminants in the country. Further research is required on quantification, implementation of monitoring and controlling system to improve the quality and safety of milk products across value chain in the regions of the country.


Introduction
Milk is a highly nutritious food containing many macroand micronutrients that are essential for the growth and maintenance of human health [1]. Milk quality and safety is a great concern in developing countries where production of milk and various milk products takes place under unhygienic conditions poor production and management practices. The common milk and dairy products contaminants are aflatoxins, toxic metals, pesticides, antibiotics and adulterants which can be controlled or/reduced by implementation of good agricultural practices and good hygienic practices at farm level and other value chain actors.
Aflatoxins are metabolic by products produced mainly by molds Aspergillus flavus and Aspergillus parasiticus. The presence of aflatoxin B 1 (AFB 1 ) in feed and the subsequent exposure of lactating animals to it leads to the contamination of milk with the hydroxyl metabolite AFM 1 [2][3][4]. AFM 1 is known to be carcinogenic and can be a potential health hazard to humans particularly to infants and children since they are more susceptible to its effect than adults [5].
Heavy metals are persistent contaminants in the environment that can cause serious environmental and health hazards [6]. This can be caused by the increase of industrial, use of new technologies and urban emissions. Milk products contain different amounts of toxic contaminants due to the grazing of livestock on contaminated land and ingestion higher levels of metals have been found in milk as a result of bioaccumulation [7,8]. These residues in milk are of particular concern even in low concentrations because milk is largely consumed by infants and children [9].
Pesticides, including the organochlorines (OCPs) are chlorine containing compounds which are found in the environment as a result of human activities [10]. OCPs are persistent and highly stable under most environmental conditions [11], which lead to high biomagnification in the food chain across a wide range of trophic levels [12]. Milk producing animals accumulate pesticides from contaminated feed and air and they are subsequently translocated and excreted through milk fat thereby consumers of milk products could be exposed to these residues [13].
Antibiotics are the primary tools for treating bacterial infectious diseases [14], and to enhance animal growth and feed efficiency in lactating cows [15]. Problems associated with antimicrobial residues in milk include increased resistance of pathogenic bacteria towards antimicrobials, risk of allergy [16], decrease acid and flavor production of butter, reduce the curdling of milk, cause improper ripening of cheeses [17].
Economically motivated adulteration is emerging risks, being addition of low cost ingredients creates an economical problem, a health risk for consumers and lowers the milk quality [18,19]. Milk is a constant target of adulteration especially in the developing world due to increased global production and consumption of milk [20]. Therefore the objective of this review is to provide information on nutritional quality, chemical contaminants and adulterants of milk products.

Materials and Methods
Review materials were collected through searching from Addis Ababa University (AAU) library, AAU website data, Science Direct, Google Scholar and Internet to gather publicly available information on the nutritional quality, chemical contaminants and adulterants of milk products. In this review method of analysis and sample sizes are mentioned in detail. The study results of articles were discussed with local and international maximum permissible limit standards of milk products. Citations are cross-check with reference lists of scientific sources.

Nutritional Quality of Milk Products
Nutritional qualities of milk and dairy products so far done in Ethiopia are presented in Table 1. The study by Dehinenet et al. [21] on raw cow's milk quality found a significant difference in all physico-chemical milk quality parameters among the districts. Breed, parity number, feeding system, farming experience and distance from dairy technology dissemination centers had significant influence on fat and protein percentages. In this work, protein level correlated positively and negatively with lactose and freezing point respectively. Dilution of milk by water (adulteration) strongly affected the nutritional quality of fresh milk. Another study by Eyassu reported average moisture, fat, protein and ash in Metata Ayib, a traditional Ethiopian fermented and spiced cottage cheese procured from a local market. A big variation in the gross nutritional composition was observed among the different Metata Ayib samples which could be attributed to the difference in the manufacturing procedure especially the spices used by the different producers. He recommended that there is a need to standardize the manufacturing procedure with respect to spices used so as to produce Metata Ayib with consistent properties.
The research accomplished by Teklemichael et al. [23] evaluated the physico-chemical properties of cow milk samples collected from dairy farms and milk vendors and reported significant difference in total solids, SNF due to poor milk handling practices in the farms and vendors. Similar study by Woldemariam and Asres [24] conducted to investigate physico-chemical qualities of different brands pasteurized milk and found no significant difference for total solids and protein contents in all pasteurized milk samples but significant difference for fat contents and total ash. The variation in physico-chemical compositions could be due to failure of regular standardization throughout the production period and they recommended to routinely examining the qualities of pasteurized milks immediately before the products are released into the market.
An assessment of physical and chemical quality of raw cow's milk produced and marketed in Shashmenie town undertaken by Teshome et al. [25]. Results showed significant difference for values of temperature, pH, titratable acidity, total solids, fat, protein, ash and lactose contents between sources of milk sample collected from collection centers, hotels, small shops and small scale producers. They conclude that nutritional composition was adequate as compared to the standard level. Evaluation of nutritional composition of raw milk conducted by Alganesh [26] from Horro cows and found similar nutritional composition of the milk in both locations and meet the acceptable standards. Significant differences for total solid, fat and protein between raw cow's milk of local and crossbred found in Walmera district [27]. They concluded that the nutritional quality of the collected raw cow's milk were within recommended levels of FAO standards. Teshome and Tesfaye [28] also reported similar finding in raw cow milk collected from three districts of Benji Maji-zone, southwestern Ethiopia.
The study by Gurmessa [29] found higher values of total solids, fat, SNF, protein, ash and lactose in milk samples from open market than household milk producers. Bruktawit and Ashenafi [30] obtained under the acceptable level of protein and fat of raw milk sample from all scale of production except freezing point, SNF and total solids in Addis Ababa sub-cities from small, medium and large farms. They suggested following hygienic practice on milk production and handling. The finding by Fikirneh et al. [31] on nutritional composition of fat, protein and SNF of raw milk obtained significantly different between breeds and among study districts. However the results of nutritional composition were adequate as compared to the standard level. Another study by Tseday and Asrat [32] observed significantly lower values for fat, SNF and water percentage of milk samples collected from consumers than producers. The poor handling practice and substandard quality of composition could be due to limited knowledge of producers and consumers on the improved hygienic handling practices. [46] The research performed by Dessalegn et al. [33] found significant difference in nutritional composition of milk samples across value chain points of Bishoftu and Akaki towns, which slightly fulfilled to the global and Ethiopian standard. The finding by Hawaz et al. [34] also showed significant difference in protein, fat, total solids and SNF of raw milk collected from milk shed through random sampling along the dairy value chain. The overall mean value of the fat and protein (5.13 and 3.51%) in the study area is higher than the Ethiopian standard (value of 3.50 and 3.20%). Alganesh [35] conducted to assess nutritional composition of milk from producers in districts of central highlands through purposive random sampling technique. The average composition of protein, total solids and ash were below Ethiopian Standard Agency. The lower average total solids might be due to the practice of adulteration and fat skimming before taking milk to collection points. The lower protein content might be due to deficiency of crude protein in the cow ration.
The finding by Anteneh et al. [36] showed that the protein, total solids and SNF of pasteurized milk marketed in Addis Ababa were below the minimum regulatory limit of the Ethiopian standards except for the fat content. The likely causes for the lower values of nutritional and physical composition were the use of low quality starting milk or addition of water. Another study by Saba et al. [37] found 32.2% positive milk samples with alcohol test while 18.8% of the samples were positive to clot on boiling test. They concluded that adequate sanitary and hygienic measures should be taken at all stages from production to consumer level. The study by Mego et al. [38] reported lower cooking butter quality due to unhygienic production and processing of butter in the study area which calls for improvement. The average moisture and ash content of butter collected from open markets of Delbo and Kucha were 18.86% and 0.16% respectively, the quality could be affected by different factors in the supply chain [39]. Belay and Janssens [40] assessed the physical and nutritional quality of raw milk collected from dairy farms which almost meets the accepted standards.
Raw milk samples collected from Sebeta, Sululta and Holeta districts contained unhygienic and poor handling practices and considered as substandard which will result in public health hazard to the consumer [41]. Of chemical composition of milk parameters, significant difference was observed in the mean fat composition among different mastitic milk [42].

Aflatoxin M 1
AFM 1 is a hydroxylated metabolite of AFB 1 that is excreted in milk in the mammary glands of both humans and lactating animals [47]. AFM 1 in milk and dairy products have been previously reviewed in different countries [1,[48][49][50][51]. The researches done on AFM 1 in milk and dairy products in Ethiopia are mentioned in Table 2. The finding by Yohannes et al. [52] showed that presence of AFM 1 in milk samples and contamination level between 0.02-0.31 µg l -1 . About 60% animal feed samples contaminated with total of AFB 1 , AFB 2 , AFG 1 and AFG 2 ranged between 4.22 and 10.54 µg l -1 in 10 samples of feed. The increase in aflatoxin caused by use of contaminated mixed daily feed by dairy farmers (mixture of wheat bran, noug cake and cotton seed cake) for cow feeding.
The research performed by Selamawit et al. [53] also studied levels of AFM 1 in milk products from industrial and local sources and found maximum mean concentration of AFM 1 in milk and yogurt from the local market relative to the industrial. However the mean values of aflatoxin contamination of all dairy types from local source were lower than the samples from industrial sources. This is because during the production processes of dairy products in the dairy industry in which raw milk collected from more than 30 individual raw milk suppliers having both higher and lower concentration of the toxin in the raw milk. Therefore they concluded that the reason for finding the maximum mean concentration of aflatoxin in the industrial dairy products is due to cross contamination during the production process.
The study conducted by Rehrahie et al. [54] reported higher AFM 1 (0.088 µg/l), from Bishoftu followed by Hawassa (0.057 µg/l) and Holetta (0.017 µg/l) sites in smallholder urban dairy producers. Likewise, concentrations of AFB 1 in different feed samples were 33.64, 20.28 and 11.07 µg/kg in Bishoftu, Holeta and Hawassa sites respectively, this indicates that AFB 1 level in the feed samples had a direct contribution for the increase in AFM 1 level in milk samples across the studied locations. Aflatoxin levels were quantified by commercial enzyme-linked immunosorbent assay (ELISA).
Another research finding by Gizachew et al. [55] on aflatoxin levels in milk samples collected from dairy farmers, milk collectors and dairy feeds (dairy farmers and processors) in the Greater Addis Ababa milk shed. They reported presence of AFM 1 in all milk samples and contamination level ranged between 0.028 and 4.98 µ/L. Overall, only nine (8.2%) out of a total of 110 milk samples contained ≤ 0.05 mg/l of AFM 1 and 29 (26.3%) exceeded 0.5 mg/l. All the feed samples were contaminated with AFB 1 (7-419 µg/kg). Out of a total of 156 feed samples collected, 41 (26.2%) of the feed samples contained AFB 1 at a level exceeding 100 mg/kg. They also found significant linear regression associations between the presence of noug cake in the feed and the levels of contamination of both AFM 1 in milk and AFB 1 in feed. Similar study by Abenet et al. [56] found mean AFM 1 levels of 11.30, 11.94, 12.88 and 29.93 ppb for Sululta, Debrebirhan, Bishoftu and Addis Ababa study sites respectively. They also reported significantly AFM 1 contamination in milk from non-grazing cows than milk from grazing cows and milk collected from Addis Ababa city.
The reduction of AFM 1 levels (57.33 and 54.04%) during lab-scale ergo (traditional Ethiopian fermented yoghurt) production was investigated by Tsige and Alemayehu [57] through determination of the residual levels of AFM 1 using ELISA. Microbiological investigation showed increasing LAB counts with incubation time. A gradual decrease in pH of the milk samples was observed during fermentation. Considering the fact that both viable and dead bacterial cells could remove AFM 1 during ergo production, the mechanism is proposed as predominantly involving non-covalent binding of the toxin with the chemical components of the bacterial cell wall.

Heavy Metals
Some of heavy metal contamination in milk products are summarized ( Table 3). The study performed by Rehrahie et al [54] found average concentration of Cd, Pb, As and Cr in cow milk samples analyzed using graphite furnace atomic absorption Spectrophotometer (GFAAS). They found that the Cd and As contents of all milk samples were within the permissible limits, whereas the concentrations of Pb in majority and Cr in all milk samples were above the permissible limits which are indicative of environmental pollution. Similarly Tassew et al. [58] studied concentrations of selected heavy metals in fresh cow's milk samples collected from four dairy farms. The elements, Co, Ni, Cd and Pb were not detected in all the milk samples. There is no significant difference in the mean concentrations of Cr and Mn between the milk samples of four farms where as that of Zn is significantly different. They reported that the detected heavy metals (Cr, Mn, Cu and Zn) can be found naturally in food. They are essential elements and thus, these metals can be found in cow's milk.
The research done by Akele et al. [59] reported heavy metal concentrations in milk samples in the following decreasing order: Zn > Mn> Cu > Cr > Cd > Pb in Chilga and Dembia and Mn> Zn > Cu > Cr in Wogera milk samples. The highest concentration in milk samples is probably due to high mineral enrichment and metal leaching (especially Cd and Pb) from coal deposits cross contamination in the chain. A study undertaken in Hawassa around Tikur-Wha textile industry on heavy metal content of milk which were reported to be unsafe [60]. This is because of contamination of livestock and irrigation areas from the industry effluents discharges. Gashu et al. [61] reported beyond acceptable level of toxic metals (Cd and Pb) in the samples of cow milk from Addis Ababa sub cities which can be a health concern for consumers. The contamination could be from environmental pollutions. Heavy metals content of traditionally prepared Ethiopian fresh butter was investigated by Shitaye et al. [46] and obtained followed in the decreasing order; Fe > Zn > Cu >Mn (mg/ Kg) in butter sample from Dirre Enchini district. Similar order was found from Ejere district, except Cu > Zn. The concentrations of the metals were in a good agreement with the international standards and indicating no risk exposure of using the butter from the investigated areas.

Organochlorine Pesticide Residue (OCP)
Few of research reports of OCP in milk and dairy product are presented in Table 4. Six persistent OCPs residues: aldrin, α-endosulfan, β-endosulfan, p, p'DDE, p, p'DDD and p, p'DDT in cow milks in Arsi, East Showa and Jimma which were selected based on information of pesticides use to control pests and malaria, reported by Deti et al. [64]. Aldrin (11.6 µg kg -1 ) detected only in one cow milk sample and α-endosulfan detected in one goat milk sample at a level of 142.1 µg kg -1 , and in one cow milk sample (47.8 µg kg -1 ) from the same region. DDE was detected in 40% of the milk samples analyzed while DDT and DDT were found in high amounts in almost all samples. The average total DDT (excluding DDD) in the samples was 328.5 µg kg -1 . Regions known for their malaria epidemics were the most contaminated with DDT residue. The accumulation pattern in both species was not clear under natural sampling. All pesticide residues found above the limit of quantification above the EU maximum residue limits (MRL). About 80% of the milk samples found to contain total DDT residue above EU MRL. Similar research by Gebremichael et al. [10] studied OCPs residues in cow's using GC-ECD and found total DDT mean levels of 0.389 µgg -1 in cow milk samples in the studied areas which are malarious, common annual spraying of DDT for malaria control, historical wide use and cheap availability in illegal markets in the area. They recommended to be carried out similar studies in other parts of the country to evaluate the breadth of the exposure. Table 5 shows level of antibiotics in milk products. The research done by Desalegn et al. [65] reported higher prevalence and amount of oxytetracycline and penicillin G residues in bovine bulk milk from Nazareth dairy farms using Delvotest SP assay and positive samples were quantified by HPLC. The antibiotic residue positive samples showed residues of oxytetracycline above the WTO/FAO/CAC established MRL of 100 µg/l were 40 (83.33%). For penicillin G, the number of samples above the MRL of 4µg/l, were 8 (16.66%). Desalegn [66] also found 34 (8.5%) positive milk samples of oxytetracycline and penicillin G residues in Debre Zeit dairy farm out of 400 samples analyzed. The major contributing factors to obtain higher prevalence and amount of antibiotic residues are lack of proper management and awareness of the people.

Antibiotic Residue
The study conducted by Yalelet et al. [67] found 36% penicillin residues out of 100 milk samples collected from healthy lactating cows of six dairy farms at Kombolcha dairy farms. Prevalence of penicillin residues in different farms ranged between 25 to 50%. Penicillin residues screened by using Delvotest SP. They recommended that coordinated nationwide surveillance of animals' by products for antibiotics residues together with determining their concentration, and initiating monitoring programmes and awareness campaigns to sensitize the populace on the dangers associated with residues in animal products in Ethiopia. The study by Tesfalem et al. [68] on antimicrobial residues in Bovine bulk milk and knowledge of farmers from Bishoftu dairy farms found high prevalence of antimicrobial residues and more than 88.5% of the farm owners' lacked proper management practices or awareness to avoid drug residues in milk and to prevent human health hazards. Takele et al. [69] conducted assessment of chemicals and drugs used for dairy and poultry farms and they recommended that laboratory investigation should be done to confirm the presence of drugs and/or chemicals in foods of animal origin (milk and its products, poultry meat, and eggs).

Adulterants in Milk Products
Few findings in adulterants of milk and dairy products are shown in Table 6. Water is the most common adulterants in milk which is often added to increase the quantity of milk for economic gain. The study by Haftu and Degnet [70] found higher water adulteration in cafeteria (2.30%) than milk producers (1.08%) in randomly collected samples using Lacto scanner. They also reported superior quality milk sample in household than cafeterias using clot on boiling tests and alcohol perception. Ayza and Yilma [71] studied patterns of milk products adulteration and reported that water, banana, vegetable oil, defatted milk and preservatives (spices and herbs) are extraneous substances added into raw milk, cheese, fresh butter, defatted milk and milk powder. They recommended further research to be done for identifying, quantifying the adulterants and the impacts in the studied area. Mekonen and Mengistu [72] evaluated water adulteration of the collected raw milk samples around Bahir Dar and reported 76.7% of milk samples collected were free from water dilution.

Conclusion and Recommendations
Milk and dairy products have poor nutritional quality, contaminated with chemical hazards and adulteration practice (water dilution) in Ethiopia. Further research is required to assess knowledge and practices of quality and safety of milk products across value chain actors in the country. Chemical contaminants should be further investigated in the regions of country. Adulterants such as formalin, sugar, salt, starch, urea and preservatives need to be studied. Intervention and awareness s creation should be conducted to improve the quality and safety of milk products.