Selenium (Se) is an essential mineral for humans and livestock. In livestock, Se is important for fertility and disease prevention [1]. Sufficient Se increases fertility rate, improves antioxidant defense systems and immunological potential [2]. Selenium-enriched diets improve milk production [3], reduce inflammatory diseases of the mammary gland [4, 5, 6], and improve growth performance [7]. Animal feeds high in Se help dairy cows to relieve oxidative stress during their transition from late gestation to early lactation due to increased metabolic activity [8]. Selenium deficiency in cattle is associated with several problems, including delayed conception, muscular degenerative disease in calves, myocardial necrosis and heart failure, immune dysfunction, increased risk of mastitis, abortion and perinatal mortality, and growth retardation in young animals [9]. It is also associated with white muscle disease which causes mortality to newborn calves and reduced productivity in growing and adult cattle [1].

We previously reported the presence of widespread human Se deficiency among the population of Ethiopia, with a strong spatial component [10]. For example, in the Amhara region, there was strong evidence of greater prevalence of Se deficiency among all demographic groups in western Amhara, but little or no deficiency was observed in eastern Amhara. This was consistent with studies of children reported previously [11]. Studies have since shown that the Se status of the Ethiopian populations is correlated with grain Se concentration of the most consumed cereal crops in Ethiopia (including maize, teff, and wheat), which in turn is linked to variation in soil and landscape properties [12, 13].

To our knowledge, there is no published information on Se status of cattle in Ethiopia. However, based on studies of soils, grains, and blood Se in people, we can expect Se deficiency to be widespread. Crop residues and stover are commonly utilized as cattle feed in Ethiopia, whose grains have already been shown to be highly variable in Se status [12, 13]. The constraining effect of Se deficiency to the livestock production is likely to be of great importance to countries like Ethiopia where the livestock sector contributes up to 80% of farmers’ incomes and 20% of the agricultural gross domestic product (GDP). In addition, Ethiopia is among the countries with the largest cattle population in the world [14]. Selenium improves the chemical composition of milk through increased Se concentration, and can reduce the concentration of saturated fatty acids, increase polyunsaturated fatty acids, and improve the organoleptic property of associated dairy products [15]. The present study analyzed Se concentration in serum of free grazing cattle, and in feed samples, using a design informed by previous studies [10, 11, 12, 13]. We hypothesized that cattle serum Se concentration in Ethiopia have similar spatial variation pattern with Se concentration in cereal grain and humans.

Statistics are reported for cows in the study set. Their serum Se concentration ranged from 14.9–167.8 $\mu$g L${}^{-1}$ (median, 41.4 $\mu$g L${}^{-1}$). Cows from East Amhara, range 24.8–167.8; median 68.4 $\mu$g L${}^{-1}$ had significantly greater serum Se concentration compared to cows from the East Amhara region, range 14.9–71.2; median 25.7 $\mu$g L${}^{-1}$, p 0.001. Overall, 79.8% of cows had serum Se concentration below the optimal threshold, 81 $\mu$g L${}^{-1}$ [20]. All of the cows from West Amhara had Se deficiency but there was relatively lower Se deficiency prevalence (62.5%) in cattle from East Amhara (Fig. 2).

Fig. 2.

Comparison of serum Se concentration among cattle from East and West Amhara region, Ethiopia. The green discs represent the original serum Se data. The yellow discs are the back-transformed mean estimates for each group (median-unbiased) with the 95% confidence interval (blue band). Note that, within each region, a small lateral random “jitter” is added so that the individual data points can be visualized.

There was marked spatially-correlated variation in cow serum Se, but the spatial dependence is very short-range after the sample origin (East or West Amhara) is included in the model. Table 1 shows the log-likelihood ratio statistic which measures the effect of sub-regions, livestock type and livestock age on cattle serum Se concentration. There is no evidence for a difference between lactating and non-lactating cows, nor for any dependence on animal age. Only the area where the sample originated is associated with Se concentration in serum.

The Se concentration of feed samples (n = 81) was in the range of 0.05–269.3 $\mu$g kg${}^{-1}$ (Table 2).

In general, feed samples from East Amhara had greater Se concentration compared to the samples from West Amhara (Fig. 3), although the sample size is small there is evidence for such a difference (p = 0.0005, Table 1b). There is also evidence for differences among the feed types (p = 0.0002, Tabl 1b) even when the East-West difference is accounted for.

Fig. 3.

Selenium concentration of feed samples from East Amhara and West Amhara region, Ethiopia. The green discs are the original Se content of the feeds. The yellow discs are the back-transformed mean estimates for each group (median-unbiased) with the 95% confidence interval (blue band). Note that, within each category, a small lateral random ”jitter” is added so that the individual data points can be visualized.

Fig. 3 shows the dominant effect of the regional trend. It is also notable that the sorghum straw appears to have a larger Se concentration than other straws. Although there were only two samples these were both much larger than for other feeds. The contrast in Fig. 4 shows only weak evidence that barley straw has a smaller Se concentration than other straws (sorghum excluded).

Fig. 4.

Comparison of Se concentration in feed samples from east Amhara and west Amhara region (the red dots are the standardized contrasts, i.e., the contrast divided by its standard error).

There is marked spatial variation in Se content of feeds, even after the regional (East-West) trend and feed differences were considered. The contrast between East and West Amhara feed Se concentration was consistent with the difference in serum Se concentration observed. This warrants regional surveys to map Se content of feeds and forage.

Results of log-likelihood ratio tests for the effects of sub-region and feed type on feed Se concentration shows that both geographical location (p = 0.0005) and type of feed (p = 0.0026) significantly influence feed Se concentration. Table 3 contains the log-likelihood ratio statistic for each comparison. The post-hoc comparisons suggest that teff straw is a richer Se source than barley straw and maize straw and (marginally) wheat straw. In addition, there is at least initial evidence that sorghum straw is the richest source of Se among feeds, although this is based on a very small sample.

The present study assessed Se concentrations in cattle serum and feed samples from West and East Amhara region where contrasting human Se status, and crop grain concentrations, had previously been established. The median serum Se concentration was 41.4 $\mu$g L${}^{-1}$. However, there was greater serum Se concentration among cattle from East Amhara compared to cattle from West Amhara. Similarly, there was greater Se concentration in feed samples from East Amhara than in samples from West Amhara. This is consistent with previously established contrasts in human Se status in these two areas [10, 11, 12, 13].

Ethiopia is one of the countries in the world with the largest cattle population [14]. However, the contribution of the livestock sector to the national economy is below its potential. Research has identified several factors hindering the development of the livestock sector in the country, including low rates of reproduction, poor feeding and healthcare practices and facilities, poor breeds, and low capital investment related to the sector [21]. The present study shows high prevalence (79.8%) of Se deficiency in cattle from the Amhara region but with spatial variation such that significantly greater deficiency prevalence was observed in cattle from West Amhara than from East Amhara. This result may suggest that the deficiency could be affecting the performance of the livestock sector in the country. Selenium deficient cattle are at higher risk of reduced growth or weight gain affecting animal fattening and meat production. The deficiency is also associated with weakened immunity in cattle which further exacerbates morbidity and low productive efficiency [22]. Selenium deficiency is also a problem of animals in other countries. For example, in Poland, Se deficiency was found in 50% of cattle. The study also reported the presence of Se deficiency variation among cattle type that 40% of cows, 80% of calves, 100% of heifers and 90% of bulls were affected [23]. However, such analysis was not possible in our study due to the unbalanced nature of the samples. Similar to results of the present study, Se deficiency was prevalent in cattle from all regions, however, there was apparent regional variation. Unlike the present study where age of cattle was not a significant factor for serum Se concentration variation, in Poland, Se deficiency was observed more frequently in young animals [23]. An earlier study in the United States also reported that about 18% of cows had severe or marginal Se deficiency and the Se status varied by geographical region and those from Southeastern states had higher rates of Se deficiency compared to cattle from other regions [20]. Se deficiency is also an important challenge to the pastoral sector in New Zealand [24].

Consistent with the distribution trend of cattle serum Se concentration, feed samples from West Amhara had lower Se concentration than East Amhara samples. Previous studies in the region [10] and the country [11] reported the presence of spatial dependence of human Se status. In addition, similar trends of spatial variability of Se concentration were observed in grain samples attributable to environmental and soil factors including soil pH, soil organic matter, temperature, rainfall and topography [12, 13].

Previously between-species variation in grain Se concentration has been reported in Ethiopia, where sorghum, finger millet and teff (Eragrostis tef (Zucc.) Trotter) had typically higher Se concentration, while barley and maize grains had the lowest Se concentration [13]. Consistent with the grain data, in the present study there was between-species variation in crop straw Se concentration, with sorghum being the richest source of Se, and teff straw containing higher Se than barley, maize, and wheat straw. This suggests that environmental controls, via the grain crops which can be cultivated at a given location, and the management effects where these are then grown by the farmer, will influence their livestock dietary supply of this essential micronutrient. Yet, this is an invisible trait to the farmer, with no role confirmed for Se in plant health. It is also likely that inter-connected geographic and species controls influence other mineral nutrients in livestock diets in this region, and these also warrant further investigation for their effects on well-being and productivity of the animals.

Results of the present study and previous studies reveal the presence of a high rate of Se deficiency in humans, plant foods and cattle from wider areas of Ethiopia. This suggests the need to communicate this aspect of deficiency to identify solutions that will increase Se nutrition for improving animal wellbeing and productivity, and human health. Crop Se fertilization helps to alleviate Se deficiency in humans [25]. This strategy may also help to improve Se status of cattle feeding on crop stovers, and the Se concentration in livestock products including milk and meat. Large-scale future studies to investigate the association between location and cattle Se deficiency in Ethiopia are important to design targeted interventions. Studies estimating the economic cost of Se deficiency to the agricultural sector in general and the livestock sector in particular is key to seek the attention of policy planners and justify the importance of the subject among other competing priorities.

The strength of this study includes generation of data on feed and serum from collocated areas, random selection of farms from villages that give spatial balance over study areas, and sufficient number of samples to observe a clear difference in Se status. Consequently, the design and sampling approach for larger surveys can now proceed on an informed basis. However, this study is based on serum analysis which reflects only recent Se exposure. In addition, this study had small area coverage. Furthermore, feed samples in the present analysis were both small in number and included only dominant feed types.

This study reveals that Se deficiency in cattle is highly prevalent though spatially variable in the Amhara region. Feed and cattle serum samples from east Amhara had significantly higher Se concentrations compared to respective samples from west Amhara. In addition, feed types were significantly associated with feed Se concentration. Selenium deficiency may affect productive efficiency of the livestock sector in Ethiopia, which should be explored together with other micronutrients, allowing interventions to enhance nutritional status in cattle to be identified. The potential for geographic variation in the nutritional quality of animal products (meat and dairy) entering the human food chain should also be studied. Further studies with larger sample size covering a larger area are needed. In addition, studies to better understand the consequences of Se deficiency to cattle productivity and the livestock sector are required.

ABS, Access and Benefit Sharing; CRM, Certified Reference Material; GDP, Gross Domestic Product; GPS, Global Positioning System; ICPMS, Inductively Coupled Plasma Mass Spectrometry; ILRI, International Livestock Research Institute; LMM, Linear Mixed Model; NIST, National Institute of Standards and Technology; Se, Selenium.

Conceptualization—KH, DG, EJMJ, SA, SG (Solomon Gizaw), SG (Samuel Gameda), ELA, RML, DBK, MRB; Data collection and supervision—KH, DG, SG (Solomon Gizaw); Sample preparation and analysis- KH, EHB, LW; Statistical analysis—RML; Original draft preparation—KH and DG; writing—review and editing— KH, DG, EJMJ, SA, SG (Solomon Gizaw), SG (Samuel Gameda), ELA, EHB, LW, RML, DBK, MRB. All authors have read and agreed to the published version of the manuscript.

Ethical approval for the study was obtained from the International Livestock Research Institute (ILRI) Institutional Research Ethics Committee (Ref. no. ILRI-IREC2018-33). Blood cattle and feed samples were collected subjected to farmers consent. Serum samples were transferred from ILRI, Ethiopia, to the University of Nottingham, UK, for analysis under a material transfer agreement and in compliance with access and benefit sharing (ABS) requirements.

The authors would like to thank the Amhara agriculture bureau, data collectors, supervisors and participating farmers. Also Kenneth Davis for assisting in the sample preparation and laboratory analysis.

Funding for part of this study was provided by the International Livestock Research Institute (ILRI) through the CGIAR Research Program on Agriculture for Nutrition and Health and by the International Maize and Wheat Improvement Center (CIMMYT) from its GIZ project Soil Protection and Rehabilitation for Food Security. This work was also supported, in part, by the Bill & Melinda Gates Foundation [INV-009129]. Under the grant conditions of the Foundation, a Creative Commons Attribution 4.0 Generic License has already been assigned to the Author Accepted Manuscript version that might arise from this submission. The funder had no role in the design, execution, analyses or interpretation of the data.

The authors declare no conflict of interest.