In recent years there has been growing interest in non-traditional feed ingredients like stevia (Stevia rebaudiana) and hemp by-products (Cannabis sativa) for dairy cow nutrition. Stevia (Stevia rebaudiana, Bertoni) is a perennial shrub native to Paraguay and Brazil, whose leaves and extracts have been widely studied and used as natural sweeteners, with additional research exploring its medicinal properties (Atteh et al., Reference Atteh, Onagbesan, Tona, Buyse, Decuypere and Geuns2011). Stevia by-products, including second-grade leaves, are rich in proteins, amino acids, lipids, carbohydrates and vitamins (Liu et al., Reference Liu, Jing, Li, Wang, Song, Zeng and Lin2024). While these nutrients are primarily relevant to human consumption, the bioactive compounds in stevia, particularly steviol glycosides, may also influence metabolic processes in ruminants (Gupta et al., Reference Gupta, Purwar, Sundaram and Rai2013; Momtazi-Borojeni et al., Reference Momtazi-Borojeni, Esmaeili, Abdollahi and Sahebkar2016; Gantelas et al., Reference Gantelas, Boukouvalas, Oikonomopoulou, Eleni and Krokida2022).
Cannabis sativa L., commonly known as industrial hemp, has been reintroduced in Greece for various industrial uses, including textiles, biofuel and building materials (Clarke and Merlin, Reference Clarke and Merlin2016; Tsaliki et al., Reference Tsaliki, Kalivas, Jankauskiene, Irakli, Cook, Grigoriadis, Panoras, Vasilakoglou and Dhima2021; Ahmed et al., Reference Ahmed, Islam, Mahmud, Sarker and Islam2022). Unlike marijuana, industrial hemp is defined by its low tetrahydrocannabinol content, making it non-psychoactive and suitable for legal cultivation (Irawan et al., Reference Irawan, Puerto-Hernandez, Ford, Busato, Ates, Cruickshank, Ranches, Estill, Trevisi and Bionaz2024a). Hemp seeds are highly nutritious, offering a source of protein, essential amino acids, beneficial fatty acids, vitamins and dietary fiber (Burton et al., Reference Burton, Andres, Cole, Cowley and Augustin2022). However, hemp also contains anti-nutritional factors such as phytic acid and tannins, which can limit nutrient absorption and should be considered when formulating animal diets (Xu et al., Reference Xu, Li, Zhao, Wang, Griffin, Li, Bean and Tilley2021; Alonso-Esteban et al., Reference Alonso-Esteban, Torija-Isasa and Sánchez-Mata2022).
Although hemp seeds have been shown to have positive effects on animal health, the use of other hemp by-products, such as hemp hay, is more complex (Irawan et al., Reference Irawan, Muchiri, Parker, van Breemen, Ates and Bionaz2024b). Not all types of hemp are approved as animal feed, and there is limited research on the safety and efficacy of these by-products for ruminants (Irawan et al., Reference Irawan, Puerto-Hernandez, Ford, Busato, Ates, Cruickshank, Ranches, Estill, Trevisi and Bionaz2024a). The use of hemp hay is particularly contentious due to the presence of cannabinoids, including cannabidiol, which must be carefully monitored to comply with regulatory limits (Parker et al., Reference Parker, Bionaz, Ford, Irawan, Trevisi and Ates2022a).
Recent life cycle assessment studies underscore the sustainability benefits of incorporating non-traditional feed ingredients, such as stevia and hemp by-products, into livestock diets. Utilizing these by-products can reduce reliance on human-edible crops, lower greenhouse gas emissions and support resource-efficient feeding practices (Barbet-Massin et al., Reference Barbet-Massin, Giuliano, Alletto, Daydé and Berger2015; Suckling et al., Reference Suckling, Morse, Murphy, Astley, Halford, Harrold, Le-Bail, Koukouna, Musinovic and Perret2023). This study evaluated the impact of adding hemp (seeds and hay) and stevia by-products to dairy cow diets on milk yield, fatty acid profile and oxidative stability. We hypothesized that these by-products would have positively effects due to their nutritional and bioactive properties. This research seeks to fill the knowledge gap regarding the use of hemp by-products in dairy cow diets, providing insights into their potential benefits and limitations, but also contributes to the broader discussion on sustainable feeding practices and their environmental benefits.
Materials and methods
The experiment was approved by the Aristotle University of Thessaloniki, and adhered to Greek Public Veterinary Service regulations. A veterinary surgeon monitored animal health, ensuring routine vaccinations against IBR, E. coli, k99 rotavirus, and coronavirus. All cows were Brucella melitensis-free, and procedures were conducted by the same team across 35 herds.
Experimental design and feeding
All experimental by-products underwent proximate analysis following the AOAC International guidelines (Official methods of analysis of AOAC International, 16th ed, 1995), which included the determination of moisture, total ash, crude protein, crude fiber and crude fat content. Additionally, the analysis of starch, sugar, neutral detergent fiber (NDF), acid detergent fiber (ADF), acid detergent lignin (ADL), total calcium, and total phosphorus content was conducted using spectrophotometric analysis (Perten DA 7250 At-line NIR Instrument). The experiment was conducted at the dairy farm of the Research Institute of Animal Science ELGO Dimitra in Paralimni, Giannitsa, northern Greece. Thirty-five purebred multiparous Holstein dairy cows with similar milk yields in early lactation were used for 70 days (10-day transition period followed by 60d experimental period) staeting at 21d post-partum. The cows were divided into five groups, namely Control, Hemp Seed, Hemp Hay, Stevia and Stevia/Hemp (a mixture of stevia leaves and hemp seed). The experimental diet included standard farm feeds (corn, barley, soybean meal, sunflower meal, corn silage, and alfalfa hay), supplemented with 7% hemp seed for the Hemp Seed group, 1 kg of hemp hay for the Hemp Hay group, 0.5% dried stevia leaves for the Stevia group, and 7% hemp seed along with 0.5% stevia leaves for the Hemp/Stevia group, all in the concentrate portion. The chemical composition of feed materials (g/kg DM) is provided in Table S1 of the Supplementary File.
We used dried hemp tops and leaves from the upper part of the Cannabis sativa plant, specifically cultivated for seed production (hemp hay), as well as second-grade stevia leaves, hemp cake and full-fat hemp seeds. The stevia leaves were sourced from ‘Stevia Hellas Coop,’ the sole Greek producer located in the Sperchios River basin, Central Greece. The cannabis by-products were supplied by Gradona, a company based in Kalochori, Thessaloniki, specializing in hemp cultivation for human nutrition. The crude protein content of the hemp hay used in the study was relatively low, reflecting its origin from a seed-producing hemp variety. All rations were formulated to be isoenergetic and isonitrogenous. Each cow was fed individually with continuous access to feed, provided ad libitum, ensuring that all cows had unrestricted access to their rations throughout the experimental period. All measurements were conducted on individual samples to ensure accuracy. Water was also available ad libitum. This approach, along with consistent feed allowance and adherence to standard commercial practices, aimed to effectively meet nutritional needs and maintain the reliability of the data collected. Dry matter intake (DMI) ranged from 25.9 to 26.7 kg/cow/day, which is consistent with the requirements for lactating dairy cows at moderate milk production levels. The daily feed intake was closely monitored to ensure that the cows were consuming the formulated amounts of each ingredient, as listed in Table 1. Given the DMI, the total intake of metabolizable energy (ME) and net energy for lactation (NEL), as well as crude protein (CP) and other nutrients, was calculated to meet the maintenance and production needs of the cows. Despite the lower CP, the inclusion of high-fiber forages (eg alfalfa hay) and supplemental protein sources (soybean and sunflower meals) supported rumen microbial activity and overall animal health. Milking occurred twice daily, with milk yield recorded at each session.
Ration ingredients and chemical composition of basal and experimental diets (% of dry matter – DM)
1 Vitamin and mineral mix contained (per kg dry matter (DM) of supplement): 9000 IU of vitamin A; 85 mg of vitamin E; 3000 IU of vitamin D3; 1.5 mg biotin; 6 mg niacin; 45 mg choline; 0.2 mg Co; 3 mg I; 100 mg Fe; 50 mg Mn; 0,45 mg Se; 150 mg Zn; 6 g of NaCl; 4 g of sulphur; 10 g of magnesium oxide; 15 g of monocalcium phosphate and 21 g of limestone.
Determination of milk yield and composition
Milk yield was recorded for each cow on days 11, 35 and 60 using the total volume measurement method, with cows identified by ear tags. At the end of the experiment, individual milk samples were analyzed for total fat, protein, lactose, total solids, solids-not-fat (SNF) and total viable counts using a MilkoScan 4000 (FOSS Electric, Denmark). Somatic cell counts were determined with a Fossomatic 5000 Basic (FOSS Electric, Denmark).
Determination of blood serum parameters
The effects of stevia and hemp by-products on health indices were evaluated by analyzing specific biochemical parameters in the blood serum, including total proteins (TP), aspartate transaminase (AST), total lactate dehydrogenase (LDH), gamma-glutamyl transferase (γ-Gt), blood urea nitrogen (BUN), creatine phosphokinase (CPK), total bilirubin (TBIL) and triglycerides (TRIGL). These assays were conducted using an automated analyzer (TARGA CLIN/CHEM Analyzer, BT 1500 Biotechnica instruments Roma, Italy). The complete blood serum analysis results are included in Table S3 of the Supplementary File.
Determination of the milk fatty acid composition
The determination of fatty acid methyl esters (FAMEs) in frozen milk samples was carried out according the method proposed by (O'Fallon et al., Reference O'Fallon, Busboom, Nelson and Gaskins2007). Detailed procedures and additional information can be found in the Supplementary File. The fatty acid composition of the samples was determined using gas chromatography.
Determination of milk oxidative stability
Lipid oxidation in milk samples, obtained at the end of the trial, was assessed using a modified spectrophotometric method based on the procedure described by Ahn et al. (Reference Ahn, Olson, Lee, Jo, Wu and Chen1998), conducted on day 1, 3 and 6 after sample collection. The concentration of malondialdehyde (MDA) in the samples was determined by comparing it to a standard calibration curve, and the results were expressed as ng MDA/mL milk. Total phenolic content (TPC) in milk samples was determined using the Folin-Ciocalteau method with modifications, following the protocol by (Li et al., Reference Li, Hosseinian, Tsopmo, Friel and Beta2009). Total antioxidant capacity (TAC) was evaluated using the method described by (Prieto et al., Reference Prieto, Pineda and Aguilar1999). Detailed procedures for these assays are provided in the Supplementary File.
Statistical analysis
A mixed linear model with two factors was employed to assess the effects of diet and time on total milk yield and oxidation. When no interactions occurred between treatment and evaluation time points, data were analyzed using one-way ANOVA. Individual cows were treated as the experimental unit. Post hoc multiple comparison tests (Tukey's test) were used to assess differences between treatment means. Assumptions of normality and homogeneity of variances were tested using Shapiro-Wilk and Levene's tests, respectively. Statistical analyses were performed using SPSS (version 28.0, IBM, provided by Aristotle University of Thessaloniki), with significance declared at P ≤ 0.05, unless stated otherwise.
Results
Milk yield and milk composition
Milk yield followed the typical lactation curve, with no significant effect of treatment and no interactions observed between treatments and time (Table 2). By the end of the experiment, no significant differences in milk yield were observed between the control and experimental groups, indeed daily milk yield remained relatively consistent across the supplemented and control groups throughout (Table 2). In addition, there was no significant effect of any supplementation of hemp and stevia byproducts on milk composition on both days 11 and 60. Data of milk composition are presented in online Supplementary Table S2.
Effect of nutritional intervention inclusion on milk yield
SEM = standard error of the mean
Blood serum parameters
The addition of hemp and stevia by-products to the diet did not significantly affect the blood serum levels on day 11. Similarly, on day 60, no significant differences were noted except for BUN levels, which showed a statistically significant decrease in all experimental groups compared to the control (P < 0.001). Additionally, TBIL levels increased significantly in the Hemp Seed and Stevia groups. A complete presentation of blood serum parameters is provided in online Supplementary Table S3.
Milk fatty acid composition
The outcomes concerning the influence of all experimental diets on the fatty acid composition of milk samples are presented in Table 3. The groups that consumed stevia produced milk with higher levels of C8:0 and C14:1 compared to the control group, although variations with other groups were not noteworthy. Moreover, hemp seed diet significantly influenced the C18:1n9c (oleic acid) and C20:4n6 (arachidonic acid) content in comparison to other groups. Overall, there was not any significant impact regarding the MUFA content of the milk between the control and experimental diets, however, a notable increase was observed in milk samples from animals fed hemp seeds compared to those fed hemp hay.
Effect of nutritional intervention on fatty acid composition of milk
SFA: Saturated Fatty Acids, MUFA: Monounsaturated Fatty Acids, PUFA: Polyunsaturated Fatty Acids
Milk oxidative stability
The inclusion of hemp and stevia by-products in the experimental groups did not impact milk oxidative stability at any point in time, evaluated at the end of the trial. However, there was clear time-related decrease in the concentration of MDA in all groups (Table 4). The addition of hemp and stevia by-products did not significantly affect the TPC and TAC values across all experimental groups (online Supplementary Table S4).
Effect of nutritional intervention inclusion on milk oxidative stability
Discussion
This study focused on the effects of hemp and stevia by-products on milk yield and composition, blood serum parameters, milk fatty acid profile and milk oxidative stability. All diets were isoenergetic and isonitrogenous, with comparable chemical compositions. The nutritional content, potential health benefits, and sustainability aspects of stevia by-products have garnered interest as feed ingredients in animal nutrition. For instance, Jiang et al. (Reference Jiang, Datsomor, Cheng, Meng, Zhan, Yang, Huang, Yan and Zhao2022) reported that replacing alfalfa hay with 6% stevia hay in dairy cow diets positively impacted milk yield. However, in our study, no significant difference in milk yield was observed in the stevia group by day 60. This outcome suggests that the level and form of stevia used, along with other dietary components, may influence the overall effectiveness of these by-products.
Hemp by-products, such as seed, hay and processing residues, are increasingly recognized as nutraceutical supplements in animal feed due to their nutrient content and potential health benefits. Karlsson et al. (Reference Karlsson, Finell and Martinsson2010), found that increasing hempseed cake levels in dairy cow diets had significant quadratic effects on milk yield, with the optimal inclusion level being 143 g/kg of dry matter. However, in our study, the inclusion of hemp hay and hemp seed did not result in a significant increase in milk yield. This discrepancy may be attributed to the relatively low levels of hemp by-products included in our diets and suggests that higher inclusion rates or different forms of hemp may be required to observe significant effects. Recent studies have demonstrated that hemp hay or by-products extracted from the hemp plant (including leaves and flowers) exhibit a fatty acid profile very similar to that of alfalfa (Parker et al., Reference Parker, Bionaz, Ford, Irawan, Trevisi and Ates2022a). As a result, these by-products do not significantly affect the milk fatty acid content or milk yield in dairy cows when included at levels of 10-20% (Irawan et al., Reference Irawan, Puerto-Hernandez, Ford, Busato, Ates, Cruickshank, Ranches, Estill, Trevisi and Bionaz2024a).
Regarding milk composition, our study found no significant adverse effects from any diet on days 11 and 60. However, the Hemp Seed group showed a notable increase in milk fat content on day 60, potentially due to the energy or fiber content of hemp seeds, aligning with studies like Klir et al. (Reference Klir, Novoselec and Antunović2019), that reported increased milk fat with hemp oil. Hemp seeds, with a protein content comparable to or greater than that of common cereals, could offer a cost-effective enhancement to dairy nutrition (Burton et al., Reference Burton, Andres, Cole, Cowley and Augustin2022).
While the inclusion of hemp and stevia by-products did not result in significant changes to most serum parameters, we observed a notable reduction in BUN levels across all experimental groups, which suggests improved protein utilization. The increase in total bilirubin levels, however, warrants further investigation as it may indicate subtle metabolic shifts associated with the dietary changes (Wang et al., Reference Wang, Tang, Yang and Gao2008, Reference Wang, Chen, Xie, Liu, Fu, Huang, Xu and Lin2020; Fallahi et al., Reference Fallahi, Ł and Opaliński2022).
Hemp by-products, rich in polyunsaturated fatty acids (PUFAs), can alter the fatty acid composition of milk, potentially improving health benefits (Șanta et al., Reference Șanta, Mierlita, Socol and Daraban2021). Mierlita (Reference Mierlita2016), found that hemp seeds enhanced the milk fatty acid profile in dairy ewes, and similar effects on c9, t11 CLA, vaccenic acid and n-3 fatty acids have been observed in cow milk (Flowers et al., Reference Flowers, Ibrahim and AbuGhazaleh2008). Stevia and its by-products, however, have not been widely documented for impacting the milk fatty acid profile. Jiang et al. (Reference Jiang, Datsomor, Cheng, Meng, Zhan, Yang, Huang, Yan and Zhao2022) noted some effects when substituting stevia hay for alfalfa, but further are missing. In our study, hemp seed diets significantly influenced C18:1n9c and C20:4n6 content, with notable increases in MUFAs compared to hemp hay. Additionally, higher levels of C8:0 and C14:1 in the milk of stevia group were noticed. The variations in fatty acids are likely due to the diet changes, with hemp seeds potentially boosting essential fatty acids in milk fat (Klir et al., Reference Klir, Novoselec and Antunović2019). Further research is needed to fully understand the effects of hemp and stevia on milk fatty acid profile through controlled feeding trials and detailed analyses.
Milk oxidative stability is influenced by unsaturated fatty acids and antioxidants. Functional fatty acids can be prone to oxidation, underscoring the need to manage lipid oxidation to preserve the nutritional quality of the milk (Șanta et al., Reference Șanta, Mierlita, Socol and Daraban2021). Mierlita (Reference Mierlita2016) found that hemp seeds improved the oxidative stability of sheep milk. Research on the impact of stevia by-products on nutritional and antioxidant properties is limited. Our study found no significant variation in MDA, a marker of lipid peroxidation, indicating stable fatty acid profiles in milk samples. We also assessed the transfer of TPC and TAC from hemp and stevia by-products into milk. No significant differences were observed, consistent with lipid oxidation findings. These results may stem from limited bioavailability, absorption efficiency and dietary interactions affecting phenolic transfer (Zeb, Reference Zeb2021; Addo, Reference Addo2022; Stobiecka et al., Reference Stobiecka, Król and Brodziak2022; Semwogerere et al., Reference Semwogerere, Chikwanha, Katiyatiya, Marufu and Mapiye2023). Further study is needed to understand how diet, genetics, and environmental factors shape antioxidant profile.
In conclusion, this study demonstrated that hemp and stevia by-products can replace a portion of traditional feed ingredients in dairy cow diets without negatively affecting lactational performance, milk composition or quality. No adverse effects were observed on milk yield, fatty acid profile or oxidative stability. However, further research is needed to optimize the inclusion levels of these by-products to improve animal performance and the production of functional dairy products. Additionally, investigating the safety of hemp, especially the potential transfer of cannabinoids like tetrahydrocannabinol and cannabidiol into milk, is essential to ensure both animal and consumer safety.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0022029925000275.
Acknowledgements
This work is part of a Phd thesis. The research was funded by Greece and the European Union (European Regional Development Fund) in context “Research—Create—Innovate” within the Operational Program (Competitiveness, Entrepreneurship and Innovation (ΕΠΑΝΕΚ) of the NSRF 2014–2020. Project Code: Τ1ΕΔΚ-04727. Acronym “GreenFeeds”.
