Farms
The study was carried out from June to September 2018 on five commercial dairy farms that housed Polish Holstein-Friesian cows in northwest Poland (Table 6). Prospective cohort data were gathered from these farms. The selection criteria for these herds included a 305-d milk yield of over 9,500 kg, a farm size of more than 80 dairy cows, housing exclusively in free-stall barns, utilization of a total mixed ration (TMR) feeding system, and a dry period (56 d).
Table 6
General farms and dairy cows’ characteristics.
Item | Farms |
1 | 2 | 3 | 4 | 5 |
General farms information |
No. cows in herds | 85 | 110 | 94 | 88 | 86 |
No. primiparous | 30 | 33 | 34 | 26 | 26 |
Average AFC1 | 734 | 754 | 790 | 774 | 789 |
Average parity no. | 3.4 | 2.8 | 2.8 | 2.6 | 2.8 |
Average 305-d milk yield (kg) | 10.021 | 10.640 | 9.987 | 10.342 | 9.875 |
General tested cows’ information |
No. tested cows | 20 | 20 | 20 | 20 | 20 |
No. primiparous | 8 | 8 | 9 | 8 | 8 |
Average AFC | 736 | 777 | 774 | 752 | 769 |
Average parity no. | 2.9 | 2.8 | 2.4 | 2.8 | 2.6 |
Average 305-d milk yield (kg) | 10.800 | 10.455 | 9.252 | 10.116 | 9.650 |
Dry period length (d) | 56 | 56 | 57 | 56 | 56 |
Predicted close-up period length (d) | 21 | 21 | 21 | 21 | 21 |
1AFC—the age at first calving. |
Animals
The part of the animal trial of heat stress has been reported also previously [16]. The dataset utilized in this study consisted of 100 clinically healthy Polish Holstein-Friesian dairy cows, selected based on parity (2.4–2.9; average 2.7) and 305-d milk yield (9,252–10,800 kg; average 10,055 kg). Twenty animals were chosen from each of the five farms participating in the study. Before dry-off, all cows were managed according to similar standard dairy farm protocols. The cows were dried off 56 d (± 1) before their expected calving date, and data recording for the prospective cohort commenced during the close-up dry period (beginning 3 weeks before calving) and continued until 150 d in milk (DIM). The calving of the first and last cows in the study occurred 42 calendar days apart. The herds maintained similar management and housing practices. During the dry period, the cows were housed in shaded free-stall barns with natural ventilation and bedding consisting of chopped straw. After calving, all cows received similar treatment, including access to cooling measures such as shade and fans or air mixers during the lactation phase. The fans or air mixers were automatically activated when the ambient temperature exceeded 21.0°C, and they maintained an air velocity of 1.8 m/s, following the guidelines of Bailey et al. [71]. The ambient photoperiod was consistent across all farms, with metal halide lighting installed over the stall areas. These lights were manually controlled to provide approximately 230 lx and a lighting duration of approximately 14 h of light and 10 h of dark. Specifically, the lights were turned on from 2000 to 0600 h. During the close-up dry period, cows were housed in group pens within the free-stall barn building, accommodating up to 10 cows/pen, with dimensions of 4 m × 12 m × 1.5 m (length × width × height). When cows displayed signs of parturition, they were transferred to individual maternity pens with straw bedding within the same building. There was a separate group for early lactation cows from 2 to 21 DIM, with an average group size of 10 cows and a stocking density of < 1 cow/lair. Lactating cows were housed in pens designed for 20 to 40 cows, featuring individual open-ended cubicles with chopped straw bedding and concrete slatted floors in the alleyways.
Representative samples of forages, including corn, grass, and alfalfa silages, were collected monthly. These samples were analyzed using the near-infrared reflectance spectroscopy method as per PN-EN 12099:2017-10, employing a FOSS DS3™ (Foss Electric, Hillerod, Denmark). The analysis covered various nutritional components, including dry matter, crude protein, neutral detergent fiber, acid detergent fiber, ether extract, and starch. This analysis was conducted consistently throughout the duration of the experiment. The nutritional value of the feed components was calculated using the nutrient content obtained through analysis. These calculations were performed using the AMTS.Cattle.Pro version 4.7 (2017, AMTS LLC, Groton, NY). The formulated diets for both close-up dry and lactating cows were adjusted monthly to meet the nutritional requirements following the guidelines set forth by the National Research Council [72]. These diets were composed primarily of silages such as corn, grass, and alfalfa, as well as concentrates such as soybean meal, rapeseed meal, barley, and triticale. A mineral–vitamin premix was also included. These diets were delivered to the cows twice daily, at 0600 and 1400 h, and were made available ad libitum. Close-up dry cows were provided with a TMR diet from 21 d prepartum until calving, with an average feed intake of 12 kg dry matter intake (DMI). After calving, all cows were transitioned to an early FRESH lactation diet up to 21 DIM, with an average feed intake of 24 kg dry matter intake (DMI), and 35 kg milk yield. Subsequently, from 22 to 150 DIM, they were fed a TMR I lactation diet, formulated to meet the nutritional requirements of dairy cows consuming 28 kg DMI and producing 42 kg of milk. The close-up diet contained a range of 65.1–76.3% forage, 14.6–15.5% CP, 35.1–39.0% NDF, and 15.9–18.5% starch, and (-45.0)–(-68.0) mEq/kg dietary cation-anion balance (DCAB) on a DM basis. The early lactation FRESH diet contained a range of 49.3–55.9% forage, 18.5–18.8% CP, 31.3–34.1% NDF, 20.7–24.4% starch, and 265–286 mEq/kg DCAB on a DM basis. The TMR I diet contained a range of 41.7–55.8% forage, 15.9–17.5% CP, 28.1–31.5% NDF, 25.5–28.4% starch, and 286–320 mEq/kg DCAB on a DM basis. Orts, which refer to leftover feed, were removed daily before the morning feeding. Throughout the experiment, water was supplied to the cows ad libitum and was accessible through self-filling troughs. The body condition score of the cows, rated on a scale of 1 (thin) to 5 (fat) according to Edmonson et al. [73] was recorded on the calving day and again 3 weeks after calving (on d 21 ± 3).
Animal health and numbers
Throughout the entire experimental period, a veterinarian conducted daily health checks on the cows to ensure their well-being. The cows received three vaccinations for rotavirus and coronavirus, administered approximately 60 and 30 days before calving, with an additional vaccination given 2 weeks after calving. For early fresh lactation cows, routine daily monitoring practices were in place. This included assessing body temperature on d 1 and d 5 after calving. Additionally, health status was monitored for various conditions, including hypocalcemia, ketosis, dystocia, retained placenta, claw health, and mastitis. Out of the initial group of 100 Polish Holstein-Friesian dairy cows, six were excluded from the study after calving, within the first 60 d post calving. This exclusion was due to health complications/disease (n = 4) and mastitis (n = 2). Data collected from these animals were included in the analysis up to the point of their removal from the study.
Environmental data
The AT and RH within the barn were continually monitored at 15-minute intervals using data loggers (WatchDog A-Series, SpecWare 9 Basic, Spectrim Technologies Inc., Aurora, USA). These data loggers had a temperature range of − 45 to 85°C with an accuracy of ± 0.6°C, and a humidity range of 0–100% with an accuracy of ± 3%. They were strategically positioned inside the building, approximately 3 m above the area where the cows were housed. This positioning aligns with the recommendations put forth by Hut et al. [15] and Lambertz et al. [26]. To assess the THI, the following formula was used, as recommended by the National Research Council [74]: THI = (1.8 × T + 32) - [(0.55 − 0.0055 × RH) × (1.8 × T − 26)]. Here, T represents the air temperature (°C), and RH stands for RH (%). In this particular study, the cows were categorized based on their THI exposure during the close-up period. The daily THI values at 21, 14, and 7 days before the calving day were calculated and these values were divided into three distinct categories: THI < 68 as a thermal comfort zone, 68–72 as mild HS, and THI > 72 as HS for dairy cows. These specific THI categories and the accompanying equation were selected because they have been used in prior dairy cow trials conducted in a temperate climate [9, 16].
Productive performance
Dairy cows were milked twice daily at 0500 and 1700 h, starting from the day of calving until 150 DIM. Their milk yield (kg) was electronically recorded. Milk samples (15 mL) were collected monthly from both morning and afternoon milkings and placed into specialized tubes with potassium dichromate as a preservative. The milk samples underwent evaluation by the Polish Federation of Cattle Breeding and Dairy Farmers. The analysis of the milk samples included assessments of protein content (%), milk urea (MU, mg/mL), fat content (%), and milk fatty acids (FAs, g/100 g of fat) using the Fourier transform infrared (FTIR) spectroscopy method according to PN-ISO 9622:2015-09. This analysis was conducted using equipment from Foss Electric (Hillerod, Denmark), specifically the MilkoScan FT 6000, along with diamond cuvettes (Foss, Hillerod, Denmark). The somatic cell count (SCC, 103 cells/mL) was determined using a Foss-o-Matic FC instrument (FOSS Electric, Hillerød, Denmark) following PN-EN 13366-2:2008. The FTIR absorption spectra encompassed 1,060 infrared frequencies (wavenumbers) spanning from 925 to 5,008/cm, representing the milk samples’ infrared light absorption characteristics. The FAs traits, expressed as g/100 g of fat, comprised four individual FAs, namely C14:0, C16:0, C18:0, and C18:1, as well as eight groups, namely saturated FAs (SFAs), unsaturated FAs (UFAs), monounsaturated FAs (MUFAs), polyunsaturated FAs (PUFAs), trans unsaturated FAs (TRANSFAs), short-chain FAs (SCFAs), medium-chain FAs (MCFAs), and long-chain FAs (LCFAs). Additionally, three ratios were calculated: the MCFA to SCFA ratio (MCFA/SCFA), the SFA to UFA ratio (SFA/UFA), and the C18 desaturation index (DI18). DI18 was determined by following the formula proposed by Komisarek et al. [63], which involves calculating the proportion of C18:1 to the sum of C18:0 and C18:1, and then multiplying the result by 100.
Fertility performance
Starting approximately 28 d (± 3) after calving, a veterinarian conducted weekly transrectal ultrasonography to assess the ovarian structures of the cows. This evaluation was performed using a color Doppler ultrasound scanner (SSD-5500, Aloka Co., Japan) equipped with a 7.5 MHz convex transducer (UST-995-7.5, Aloka Co., Japan). The veterinarian scanned each ovary from multiple angles to thoroughly assess the position of the structures. The confirmation of the first estrus post-calving was based on the presence of a visible corpus luteum (dominant follicle measuring ≥ 10 mm in diameter) in either ovary. Subsequently, the cows were examined weekly until the second estrus was confirmed. Day 1 of the estrus cycle was designated as the day when the new estrus was confirmed. This day marked the period between the disappearance of a dominant follicle measuring ≥ 10 mm in diameter and the appearance of a new corpus luteum in the same position. Cows exhibiting signs of estrus were artificially inseminated (AI). Pregnancy was diagnosed by transrectal ultrasonography on d 30 (± 3) after AI. The fertility of dairy cows was evaluated based on several indices such as the first estrus postpartum (number of days from calving to the first estrus), calving interval (number of days from calving to the first service), insemination period service (days from the first AI service to conception), services per conception (number of AI services required for conception), days open (number of days from calving to the next fertilization), and inter-calving period (number of days from one calving to the next calving).
Blood sampling and analyses
Blood samples were collected from each cow approximately 4 ± 0.5 h after morning feeding, precisely 3 weeks after calving (on d 21 ± 3) during standard veterinary procedures. The samples were obtained from the tail vein using a 10 mL vacutainer designed for serum collection (KABE, Poznan, Poland). Following collection, the vacutainers containing the blood samples were subjected to centrifugation at 3,000×g for 15 min at 4°C. This process allowed for the isolation of serum, which was then stored at − 20°C until further analysis. The serum samples were utilized to determine the concentrations of insulin, insulin-like growth factor-I (IGF-I), NEFA, adiponectin (ApN), leptin (LEP), follicle-stimulating hormone (FSH), luteinizing hormone (LH), TNF-α, IL-1, and IL-6 using bovine ELISA kits (Shanghai Sunredbio Technology, China). All assays were performed in duplicate, and the absorbance values were measured at 450 nm using a microplate spectrophotometer (Synergy 2, BioTek, BIOKOM, Warsaw, Poland). Specifications of bovine ELISA kits from Shanghai Sunredbio Technology (China) used for determining hormones, cytokines, and metabolic blood indices were reported previously [16]. Additionally, the concentration of β-hydroxybutyrate (BHB) was measured using a bovine kit manufactured by Pointe Scientific (Warsaw, Poland). The BHB assay range was 0.05 to 6.9 mmol/L (catalog no. H7587), and absorbance values were read at 340 nm using the colorimetric method and the same microplate spectrophotometer (Synergy 2, BioTek, BIOKOM, Warsaw, Poland). Quality control measures were implemented to ensure the reliability of the results, with both interassay and intraassay variation kept within a coefficient of variation of ≤ 5% for all blood variables.
Statistical analyses
Before analysis, all data underwent normality screening using PROC UNIVARIATE in SAS version 9.4, SAS Institute. [75]. Fertility parameters, such as service per conception and AI period service, were subjected to a logistic transformation function before statistical analysis. The explanatory variables for THI during the close-up dry period (− 21, − 14, and − 7 d) were divided into three categories, which were consistent for the analysis of productive performance, fertility, and immunometabolic blood indices. The effect of HS, defined by the THI during the close-up period, on fertility, productive performance, and immunometabolic blood indices was determined using the MIXED procedure of SAS under the models described in the equations. The best covariance structure was determined using the lowest Bayesian information criterion fit statistic level, and the autoregressive variance–covariance structure was selected accordingly.
Productive performance was analyzed using the following model: Yijklmn = µ + Li + Hj + Bk + β1afcl + β2dlm + eijklmn, where Yijklmn is the value of the dependent variable, µ is the overall mean, Li is the fixed effect of parity of dam (i = 1, 2), Hj is the fixed effect of the farm (j = 1, 2, 3, 4, 5), Bk is the explanatory variables (k = effect of the category of THI on the − 21, −14, and − 7 d during the close-up period; listed in Table 7), β1 and β2 are the partial linear regression coefficients, afcl is age at first calving, dlm is DIM, and eijklmn is the random error. The fertility indices were analyzed using the same statistical model but without the second partial regression coefficient (β2dlm). The immunometabolic blood indices were analyzed using the following model: Yijklmn = µ + Li + Hj + Bk + β1afcl + β2dsm + eijklmn, where Yijklmn is the value of the dependent variable, µ is the overall mean, Li is the fixed effect of parity of dam (i = 1, 2), Hj is the fixed effect of the farm (j = 1, 2, 3, 4, 5), Bk is the explanatory variables (k = effect of the category of THI on the − 21, −14, and − 7 d during the close-up period), β1 and β2 are the partial linear regression coefficients, afcl is the age at first calving, dsm is the day of blood sampling, and eijklmn is the random error.
Table 7
Descriptive statistics of evaluated parameters (explanatory variables) during the close-up period of dairy cows.
Item1 | n | Mean | Minimum | Maximum | SEM |
THI − 21 d | 100 | 64.1 | 51.6 | 77.2 | 2.68 |
THI − 14 d | 100 | 66.6 | 52.5 | 79.4 | 2.53 |
THI − 7 d | 100 | 67.3 | 52.9 | 78.1 | 2.40 |
1THI − 21 d = temperature-humidity index calculated on the − 21 d before the calving day and divided into three categories: THI < 68 as a thermal comfort zone (n = 31), 68–72 as mild heat stress (n = 35), and THI > 72 as heat stress for dairy cows (n = 34); THI − 14 d = temperature-humidity index calculated on the − 14 d before the calving day, and divided into three categories: THI < 68 as a thermal comfort zone (n = 31), 68–72 as mild heat stress (n = 31), and THI > 72 as heat stress for dairy cows (n = 38); THI − 7 d = temperature-humidity index calculated on the − 7 d before the calving day, and divided into three categories: THI < 68 as a thermal comfort zone (n = 31), 68–72 as mild heat stress (n = 31), and THI > 72 as heat stress for dairy cows (n = 38). |
When the explanatory variables (categories of THI on − 21, −14, and − 7 d during the close-up period) showed significance, individual comparisons were conducted using Duncan's adjustment. Statistical significance was declared when P ≤ 0.05, and trends were noted when 0.05 < P ≤ 0.1. Pearson correlation coefficients were calculated using the PROC CORR procedure in SAS.
Ethics statement
All procedures for the study were approved and performed following the guidelines of the Polish Council for Animal Care (Act on the Protection of Animals Used for Scientific Purpose in Poland), which complies with the EU directive (no. 2010/63/EU) for the protection of animals used for scientific purposes [76]. These practices are standard for animal health assessment and monitoring; in particular, blood samples were collected during standard veterinary activities, and all procedures were approved by the appropriate authorities. All animal procedures and methods are reported in accordance with ARRIVE guidelines.