Oxidative Stress Biomarkers and Metabolic Parameters in Healthy Holstein Dairy Cows and Cows With Left Displacement Abomasum During the Transitional Period

ABSTRACT

Background

During the transitional period, dairy cows experience oxidative stress and are more susceptible to diseases, including left displacement of the abomasum (LDA).

Objectives

This study aimed to compare oxidative stress biomarker levels in cows with LDA to those in healthy conditions and investigate the associations between predictive metabolites linked to LDA and oxidative stress biomarkers.

Methods

In this case–control study, 400 healthy multiparous Holstein cows were matched for lactation number, milk production and calving date. Blood samples were collected at four time points: 21 and 7 days before, as well as 7 and 21 days after parturition from all animals. During the observation period, seven cows diagnosed with LDA in the main population, and seven healthy cows were randomly selected as controls for the comparison of oxidative stress, liver enzymes and metabolic parameters. Analysis of blood parameters utilized repeated measures ANOVA, and the degree of relationship between oxidative stress biomarkers and other measured parameters was assessed using Pearson correlations.

Results

The LDA group exhibited significantly higher levels of urea, blood urea nitrogen (BUN), gamma‐glutamyl transferase, glucose, cholesterol, triglyceride (TAG), β‐hydroxybutyric acid (BHBA), aspartate aminotransferase (AST), sorbitol dehydrogenase, serum amyloid A (SAA), chloride, sodium, potassium, total antioxidant capacity (TAC) and malondialdehyde (MDA) compared to the control cows (p < 0.05). Positive correlations were observed among BUN, glucose, TB, AST, SAA, BHBA, TAG and MDA. Conversely, these parameters displayed negative correlations with TAC. Negative correlations were found among chloride, sodium, potassium, calcium, phosphorus and MDA, whereas positive correlations were observed with TAC.

Conclusions

These findings highlight the elevated level of oxidative stress and compromised antioxidant defence in cows with LDA and the intricate interplay among oxidative stress, metabolic parameters and liver enzymes.

During the transitional period, dairy cows exhibit heightened susceptibility to diseases, including left displaced abomasum (LDA). A significant shift in oxidative stress, metabolic parameters and hepatic enzyme profiles in LDA‐affected cows was observed. These observations underscore an escalated oxidative stress status and compromised antioxidant defence mechanisms in cows with LDA.

1. Introduction

This study aimed to compare oxidative stress biomarkers, metabolic parameters and liver enzymes in dairy cows with left displacement of abomasum (LDA) versus healthy cows. Additionally, it aimed to establish correlations between predictive metabolites linked to LDA and oxidative stress biomarkers during the transitional period in dairy cows.

The transitional period, lasting 3 weeks before and after parturition, is challenging for dairy cows (Drackley, 1999) with a higher susceptibility to various diseases, including LDA, during this time (Geishauser 1995; Mallard et al. 1998) due to several factors, including hypocalcaemia, metritis and negative energy balance (NEB), all of which contribute to the onset of LDA. NEB is characterized by metabolic changes, such as changes in blood glucose, ketonaemia, increased concentrations of non‐esterified fatty acids (NEFAs) and β‐hydroxybutyric acid (BHBA), and the accumulation of triacylglycerol in the liver that significantly contribute to the pathogenesis of LDA (Shaver. 1997; Herdt. 2000). Dairy cows often experience oxidative stress during the transition period due to metabolic demands and pathogenic challenges (Sordillo et al. 2009; Sharma et al. 2011; Sordillo and Mavangira 2014; Sabzikar, Mohri, and Seifi 2023), which can cause several diseases, including LDA (Constable et al. 2016).

Recent studies highlight ruminants’ vulnerability to oxidative stress in challenging situations, including weaning (Kazemi et al. 2022), transportation (Mohamed et al. 2023), parturition (Sabzikar, Mohri, and Seifi 2023) and environmental stress (Sesay et al. 2023). The growing interest in oxidative status in ruminant medicine emphasizes the need for reliable methods to assess oxidative stress. However, limited studies have explored the status of oxidative stress biomarkers in cows with LDA (Castillo et al. 2006; Sordillo 2005; Fiore et al. 2019). High oxygen demands during metabolic challenges increase reactive oxygen species (ROS) generation. This ROS–antioxidant imbalance during the periparturient period causes elevated oxidative stress in dairy cows (Sordillo & Aitken 2009). Studies confirm that oxidative stress impairs immune and inflammatory responses in farm animals, including dairy cows, during metabolic challenges like the transitional period, increasing their vulnerability to health disorders (Sordillo 2005; Wilde 2006; Sharma et al. 2011; Abuelo et al. 2013) and dietary supplementation has been utilized as an approach to decrease the presence of detrimental metabolites (Mirzaei et al. 2020; Ayemele et al. 2021; Kazemi et al. 2022). Haematological biomarkers serve as diagnostic indicators for assessing abdominal disorders in dairy cows, reflecting metabolic state, stress levels, injuries and inflammation (de Cardoso et al. 2008). Previous studies used biomarkers for identifying LDA, such as acute‐phase response, oxidative stress and hepatic function (Maden et al. 2012), but the correlation among changes in oxidative stress biomarkers, metabolic parameters and liver enzymes in LDA cows remains unexplored.

A comprehensive analysis of oxidative stress biomarkers in LDA‐affected cows and healthy cows, along with exploring LDA‐associated metabolites, provides insights into disrupted metabolites and oxidative stress. This approach enhances the understanding of metabolic dysfunction, oxidative stress and LDA development in dairy cows. Valuable insights aid the development of strategies to improve nutrition and reduce oxidative stress–related metabolites in cows with LDA. Therefore, the objective of this study is to compare the oxidative stress biomarkers, liver enzymes and metabolic parameters in healthy cows versus LDA cows and the correlation of these factors with oxidative stress biomarkers in a field condition during the transitional period in dairy cows.

2. Materials and Methods

2.1. Animals, Housing and Diet

The dairy farm housed approximately 1200 lactating cows, which were subjected to three‐time milking sessions in a parallel parlour. Cows were housed in ventilated free‐stall barns with sand bedding. Each pen had two rows of free‐stalls, with sprinklers over the feed‐bunk and fans over the feed‐bunk and stalls. Prepartum cows stayed in separate pens with two rows of stalls in a naturally ventilated, sand‐bedded free‐stall barn. When signs of calving were observed, the cows were immediately moved to a straw‐bedded loose‐housing pen. Cows were fed a formulated total mixed ration (TMR) meeting nutritional requirements based on animal categories, milk yield and NRC 2001 guidelines. TMR was offered twice daily. Leftover feed was removed and fresh portions served at 9 AM and 7 PM. Mean dry matter intake was recorded for the observational cow groups. Prepartum diets met or exceeded nutrient requirements for non‐lactating Holstein cows weighing 725 kg, with 0.6 kg/day conceptus weight gain and 10 kg/day DMI (NRC, 2001). Postpartum diets met or exceeded nutrient requirements for lactating Holstein cows weighing 650 kg, producing 45 kg of 3.5% FCM (NRC, 2001). The primary components of the postpartum diets and the composition of the diet for both groups are presented in Tables 1 and 2, respectively.

TABLE 1.

Ingredients and nutrient composition of diet administered during transitional period.

Ingredients (% of DM)	Close up	Fresh
Alfalfa hay	18.30	19.14
Corn silage	23.49	25.60
Wheat straw	20.21	0
Barley grain	3.83	12.52
Corn grain	7.66	13.77
Fresh mix	0	1.66
DCAD premix	3.40	0
Lact min‐vit premix a	2.55	2.22
Trans min‐vit premix b	2.55	2.22
Soybean meal	5.24	8.62
Canola meal	5.99	5.38
Sesame meal	2.99	3.91
Whole cotton seed	3.74	4.89

^a Lactation vitamin premix.

^b Transitional vitamin premix.

TABLE 2.

Composition of the total mixed ration (TMR) diet administered during transitional period.

Nutrients (% of DM)	Close up	Fresh
DM, % as fed	50.8	51.5
Crude protein	15.1	17.8
RUP a	32.3	34.5
NEL, Mcal/kg b	1.46	1.64
ADF c	26	19.3
NDF d	41	28.8
NFC e	35.5	41.2
Starch	19.8	26.5
Ether extract	3.1	3.8
Ash	9.5	7.9
Calcium	1.15	1.23
Phosphorous	0.45	0.53
Magnesium	0.46	0.41
Potassium	1.35	1.24
Sodium	0.16	0.38
Chloride	1.39	0.27
Sulphur	0.25	0.22
DCAD, meq/kg	−118	210

^a Rumen undegradable protein.

^b Net energy.

^c Acid detergent fibre.

^d Neutral detergent fibre.

^e Non‐fibre carbohydrates.

2.2. Study Design

In this case–control study, 400 multiparous Holstein cows (3rd to 5th lactation) coordinated in milk production and calving date that calved during winter were randomly selected from a commercial dairy farm located in the south–middle region of Iran. Cows with an initial body weight (BW) of 652.5 ± 42.3 kg (mean ± SD), body condition score (BCS) of 3.5 ± 0.25 (mean ± SD) and the average milk production of 38.7 ± 6.4 kg/day (mean ± SD) in the previous lactation were selected for a 42‐day trial consisting of four samplings in a completely random design. First and second blood samples were taken 3 and 1 week before parturition. Postpartum blood samples were taken 1 and 3 weeks after calving. Among the sampled cows, 11 had LDA. All 400 cows underwent the 4 sampling times. However, it's important to note that 300 of these cows received supplementation challenges that were distinct from the ones applied in the current experiment. The details of these supplementation challenges will be addressed in separate scientific manuscripts. Four cows were excluded due to concurrent diseases (retained placenta, metritis and lameness). One cow was removed as initial exploratory laparotomy did not confirm LDA, making surgery inappropriate. For further blood analysis, blood samples from the 7 cows with confirmed LDA were utilized (LDA cows). To establish a comparison group, 7 cows (lactation number = 3rd to 5th lactation) were randomly selected from the remaining 89 cows that did not have LDA (healthy cows). The 300 remaining cows from the pool of 400 cows were employed in other simultaneous studies involving distinct challenges. Consequently, including them in our randomization pool was deemed inappropriate. The inclusion criteria for the selected healthy cows were as follows: They had the same history of pregnancy synchronization time, approximately the same amount of milk production in the previous lactation, similar BCS at drying off, the same length of dry period and comparable calving ease. Additionally, all the animals underwent a thorough clinical examination at drying off and immediately after calving to ensure the absence of any underlying diseases or health issues. Characteristics of the included multiparous Holstein cows assigned to the two groups of study are presented in Table 3.

TABLE 3.

Characteristics of the included multiparous Holstein cows randomly assigned to the two groups of study, including control (n = 7) and left displacement of the abomasum (LDA) (n = 7) at time of enrolment.

Item	Control (Mean ± SE)	LDA (Mean ± SE)	p value
Body weight a	640.11 ± 11.19	639.14 ± 11.15	0.95
Body condition score b	3.24 ± 0.77	3.25 ± 0.77	0.94
Milk yield in previous lactation c	11,808.82 ± 6.16	11,801.02 ± 6.57	0.39
Dry period length d	61 ± 1.06	59.85 ± 0.93	0.93
Close up period length d	30.71 ± 0.91	30.42 ± 0.75	0.81
Calving easee	1.14 ± 1.14	1.14 ± 1.14	1

^a Body weight expressed as Kg.

^b The body condition score is categorized on a scale from 1 to 5 according to the Pennsylvania State University method.

^c Milk production in previous lactation for a 305‐day lactation period, expressed in Kg.

^d The duration of time, measured in days, during which the animal resided in the specified pen.

^e Calving ease is categorized on a scale from 1 to 4, with 1 indicating unassisted calving and 4 representing calving requiring cesarean section or fetotomy.

2.3. Disease Definition

Cows with decreased appetite upon returning from the milking parlour and a reduction in daily milk yield over 25% underwent clinical examination. LDA was identified by metallic resonance during percussion and auscultation of the left abdominal regions. Ultrasound examination confirmed the diagnosis by visualizing reverberation artefacts. Abomasocentesis was conducted through the left abdominal wall's middle third, revealing fluid with a pH around 2. Laparoscopic examination further confirmed the diagnosis of LDA. All suspected cases of LDA were finally confirmed by exploratory laparotomy and correction of the displaced abomasum.

2.4. Blood Sampling and Measurement of Blood Parameters

Blood samples were collected at specified intervals relative to calving, which included time points of 21 days before calving, 7 days before calving, 7 days post‐calving and 21 days post‐calving. The collection was performed by the study personnel who were blinded to the clinical status of the animals, ensuring double‐blinding conditions for the sampling process. The samples were obtained from the jugular vein before the morning feeding at 9 AM using an 18‐G, 1‐in long needle and were then placed into plain evacuated tubes. These tubes were immediately chilled on ice until subjected to centrifugation at 1200 × g for 15 min at 4°C to separate serum. Subsequently, the serum was divided into microcentrifuge tubes and preserved at a temperature of −32°C until further analysis. The analysis was performed by laboratory personnel who were also unaware of the animal grouping, maintaining the blinding sampling condition throughout the study.

Serum biochemical parameters, including blood urea nitrogen (BUN), total bilirubin, glucose, cholesterol, triglyceride (TAG), total protein, albumin, aspartate aminotransferase (AST), gamma‐glutamyl transferase (GGT), calcium (Ca), phosphorous (P) and chloride (Cl), were measured using standard methods and commercial kits (Pars Azmoon Co., Tehran, Iran) and a biochemical auto analyser (Alpha Classic AT++, Sanjesh, Iran). Serum concentrations of sodium (Na) and potassium (K) were measured using a flame photometer apparatus (Fater Electron Company, Tehran, Iran). NEFA was measured using a commercial ELISA kit (ZellBio GmbH kit, Germany, ZB‐FFA96A). BHBA was measured using a commercial ELISA kit (ZellBio GmbH kit, Germany, ZB‐11497‐H9648). Serum sorbitol dehydrogenase (SDH) was measured using a commercial ELISA kit (ZellBio GmbH kit, Germany, ZB‐10807C‐H9648). Serum amyloid A (SAA) was measured using a bovine solid‐phase sandwich ELISA method (Shanghai Crystal Day Biotech Company, China): SAA intra‐assay: CV < 8%, inter‐assay: CV < 10%, SAA; and sensitivity 0.156 pg/mL. A commercial kit (ZellBio GmbH kit, Germany) was used to determine the total antioxidant capacity (TAC) level. The colour product of the chromogenic substrate (tetramethyl benzidine) emerged at the ending phase. The difference in colour was calculated calorimetrically using a spectrophotometer (Jenway 6300 Spectrophotometer, UK) at 450 nm and represented as mmol/L. This method can determine TAC with 0.1 mM sensitivity (100 µmol/L). The intra‐ and inter‐assay CVs were below 13.4% and 4.2%, respectively. An assay kit purchased from ZellBio GmbH (Germany) measured malondialdehyde (MDA) (µmol/L; Cat. no. ZB‐MDA96A). In this kit, MDA is measured based on its reaction with thiobarbituric acid in an acidic condition at high temperature. The colour complex was measured colourimetrically at 535 nm. The assay kit sensitivity was 0.1 µM (inter‐assay CV: 5.8%) for MDA.

2.5. Statistical Analysis

All data were analysed using SAS version 9.3 (SAS/STAT, SAS Inst. Inc., Cary, NC, USA). A linear mixed model was designed to analyse blood metabolites using repeated measures ANOVA with the MIXED procedure of SAS. Initial models had five fixed effects, including treatment, sampling time, interaction of sampling time and treatment, lactation number and BCSs of the animals at the beginning of the transitional period. Treatment and sampling time were constant in all models. The interaction of treatment and sampling time, lactation number and BCSs were included only if the p value was ≥0.1. Treatment was the experimental unit, with the animals as the random effect nested within treatment, and sampling time was specified as the repeated measure. Due to unequal sampling intervals, three variance–covariance structures (compound symmetry, compound symmetry heterogenous and spatial‐power structure) were tested. The compound symmetry structure, which had the lowest Schwarz's Bayesian information criterion, was selected and used in the models (Littell, Henry, and Ammerman 1998). Initial measurements (first sampling time) served as covariates in data analysis. The Tukey–Kramer method was used to calculate differences in least square means (LSM) for multiple comparisons, comparing different levels of independent variables with the outcome of interest. The results, along with their corresponding standard errors of the mean, were presented. Degrees of freedom were estimated by using the Kenward–Roger method in the model statement (Littell, Henry, and Ammerman 1998). To determine the degree of relationship between oxidative stress biomarkers and other measured parameters, correlation analysis was performed using Pearson correlations and the CORR procedure of SAS. Results of the correlation analysis between estimated parameters were presented in Table 3. Residual distribution for each variable was evaluated for normality and homoscedasticity of the data. Responses deviating from the assumptions of normality (GGT and total bilirubin) underwent power transformation following the Box–Cox procedure (Box and Cox 1964), utilizing a SAS macro for mixed models developed by Piepho (2009). The LSM and corresponding standard errors of the means were back‐transformed in accordance with the methodology proposed by Jørgensen and Pedersen (1997) for the purpose of result presentation. According to the Pearson residual method, the rest of the residuals for other variables are presented in the normal distribution. p ≤ 0.05 was considered significant, and trends were declared at 0.05 < p ≤ 0.10.

3. Results

3.1. Oxidative Stress Biomarkers

A significant effect of sampling time (p < 0.001) and treatment (p = 0.03) on the average concentration of TAC was observed. Furthermore, the average concentration of TAC was (p < 0.001) affected by the interaction of sampling time and treatment (Figure 1). Average concentration of TAC across different sampling times in two different treatments is presented in Table 6. The average concentration of TAC was lower in cows with LDA (mean = 0.80 mmol/L, 95% confidence interval = 0.72, 0.88) compared to the healthy cows (mean = 0.93 mmol/L, 95% confidence interval = 0.85, 1.01).

FIGURE 1.

Changes in total antioxidant capacity (TAC) and malondialdehyde concentration in Holstein cows in different experimental groups (LDA‐affected cows and control group). The first, second, third, and fourth sampling times were 3 and 1 week before calving and 1 and 3 weeks after calving, respectively. The TAC concentration exhibits significant differences between the control group and LDA group in the fourth sampling time (p < 0.05). As for MDA concentration, there is a tendency for difference (0.05 < p ≤ 0.10) in the second sampling time and a significant difference (p < 0.05) in the third and fourth sampling times when comparing the LDA group with the control group. LDA, left displacement of the abomasum.

TABLE 6.

Average serum concentration of measured parameters in the healthy group and the left displacement of the abomasum (LDA) group in each sampling time.

	Control group (mean ± SE)				LDA group (mean ± SE)
Variable	1	2	3	4	1	2	3	4
Urea (mg/dL)	36.14 ± 5.03	50.29 ± 5.03	47.14 ± 5.03a	65.86 ± 5.03	48.00 ± 5.03	53.43 ± 5.03	70.00 ± 5.03b	82.86 ± 5.03
BUN (mg/dL)	16.89 ± 2.35	23.45 ± 2.35	22.03 ± 2.35a	30.77 ± 2.35	22.43 ± 2.35	24.97 ± 2.35	32.71 ± 2.35b	38.72 ± 2.35
GGT (U/L)	18.80 ± 2.11	26.86 ± 2.11	26.23 ± 2.11a	18.90 ± 2.11	19.34 ± 2.11	23.64 ± 2.11	38.78 ± 2.11b	19.34 ± 2.11
TB (g/dL)	41.85 ± 2.96	26.65 ± 2.96	26.19 ± 2.96	24.47 ± 2.96	47.12 ± 2.96	27.64 ± 2.96	23.06 ± 2.96	20.28 ± 2.96
AL (g/dL)	27.26 ± 1.02	27.59 ± 1.02	24.54 ± 1.02A	26.26 ± 1.02A	25.25 ± 1.02	27.471 ± 1.02	28.97 ± 1.02B	30.64 ± 1.02B
TP (g/dL)	7.51 ± 0.18	7.49 ± 0.18	7.00 ± 0.18	7.29 ± 0.18	7.13 ± 0.18	7.14 ± 0.18	7.06 ± 0.18	7.47 ± 0.18
Gl (g/dL)	4.79 ± 0.21	4.76 ± 0.21	4.55 ± 0.21	4.66 ± 0.21	4.60 ± 0.20	4.40 ± 0.20	4.16 ± 0.20	4.41 ± 0.20
Al/Gl	5.77 ± 0.47	5.84 ± 0.47	5.55 ± 0.47	5.66 ± 0.47	5.62 ± 0.47	6.45 ± 0.47	7.14 ± 0.47	7.05 ± 0.47
Glu (mg/dL)	56.93 ± 10.29	82.21 ± 10.29	107.86 ± 10.29	43.14 ± 10.29a	65.43 ± 10.29	80.71 ± 10.29	109.43 ± 10.29	128.14 ± 10.29b
Chol (mg/dL)	2.85 ± 0.26	3.27 ± 0.26	2.72 ± 0.26	3.38 ± 0.26	2.33 ± 0.26	2.64 ± 0.26	2.87± 0.26	3.24 ± 0.26
TG (mg/dL)	19.31 ± 2.35	23.93 ± 2.35	18.21 ± 2.35a	13.33 ± 2.35a	17.71 ± 2.35	28.81 ± 2.35	30.00 ± 2.35b	36.54 ± 2.35b
NEFA (mmol/L)	0.51 ± 0.02A	0.48 ± 0.02	0.49 ± 0.02a	0.51 ± 0.02a	0.53 ± 0.02B	0.49 ± 0.02	0.56 ± 0.02b	0.55 ± 0.02b
BHBA (mmol/L)	0.71 ± 0.05a	0.84 ± 0.05	0.73 ± 0.05a	0.77 ± 0.05a	0.77 ± 0.05b	0.86 ± 0.05	1.43 ± 0.05b	1.30 ± 0.05b
AST (U/L)	44.00 ± 11.32	102.26 ± 11.32	102.87 ± 11.32a	62.50 ± 11.32a	59.50 ± 11.32	80.14 ± 11.32	165.00 ± 11.32b	124.86 ± 11.32b
SDH (U/L)	11.51 ± 0.78	11.39 ± 0.78	8.26 ± 0.78a	10.70 ± 0.78	10.11 ± 0.78	12.09 ± 0.78	13.23 ± 0.78b	13.21 ± 0.78
SAA (µg/mL)	2.72 ± 0.41	3.19 ± 0.41	1.92 ± 0.41a	2.61 ± 0.41a	3.56 ± 0.41	4.71 ± 0.41	5.47 ± 0.41b	5.98 ± 0.41b
Cl (mmol/L)	85.00 ± 2.26	80.36 ± 2.26	79.86 ± 2.26	84.29 ± 2.26a	91.63 ± 2.26	73.83 ± 2.26	71.80 ± 2.26	69.07 ± 2.26b
Ca (mg/dL)	7.76 ± 0.21	6.90 ± 0.21	6.86 ± 0.21	8.09 ± 0.21a	8.44 ± 0.21	6.70 ± 0.21	6.86 ± 0.21	7.10 ± 0.21b
P (mg/dL)	5.25 ± 0.33a	5.46 ± 0.33	4.93 ± 0.33	5.57 ± 0.33A	6.97 ± 0.33b	5.36 ± 0.33	4.63 ± 0.33	4.19 ± 0.33B
Na (mg/dL)	122.86 ± 2.03	119.57 ± 2.03a	117.86 ± 2.03a	126.00 ± 2.03a	116.86 ± 2.03	102.29 ± 2.03b	99.29 ± 2.03b	98.43 ± 2.03b
K (mg/dL)	4.10 ± 0.25	4.13 ± 0.25	3.94 ± 0.25	5.04 ± 0.25a	3.87 ± 0.25	3.30 ± 0.25	3.36 ± 0.25	2.89 ± 0.25b
TAC (mmol/L)	0.99 ± 0.06	0.96 ± 0.06	0.78 ± 0.06	0.99 ± 0.06a	1.15 ± 0.06	0.78 ± 0.06	0.66 ± 0.06	0.61 ± 0.06b
MDA (nmol/mL)	1.78 ± 0.24	2.47 ± 0.24A	3.33 ± 0.24a	1.83 ± 0.24a	1.86 ± 0.24	3.45 ± 0.24B	4.74 ± 0.24b	4.99 ± 0.24b

Note: Interaction of treatment and sampling time. Within the same sampling times, values with different superscript (a,b) differ (p ≤ 0.05). Within the sampling times, values with different superscript (A,B) differ (0.05 < p ≤ 0.10).

Abbreviations: AL/Gl, albumin/globulin ratio; AL, albumin; AST, aspartate aminotransferase; BHBA, β‐hydroxybutyric acid; Ca, calcium; Chol, cholesterol; Cl, chloride; Gl, globulin; Glu, glucose; K, potassium; MDA, malondialdehyde; Na, sodium; NEFA, non‐esterified fatty acids; P, phosphorous; SAA, serum amyloid A; SDH, succinate dehydrogenase; TAC, total antioxidant capacity; TB, total bilirubin; TG, triglyceride; TP, total protein.

A significant effect of sampling time (p < 0.001) and treatment (p < 0.001) on the average concentration of MDA was observed. Furthermore, the average concentration of MDA was (p < 0.001) affected by the interaction of sampling time and treatment (Figure 1). Average concentration of MDA across different sampling times in two different treatments is presented in Table 6. The average concentration of MDA was higher in control cows (mean = 3.76 nmol/mL, 95% confidence interval = 3.39, 4.13) compared to the cows with LDA (mean = 2.36 nmol/mL, 95% confidence interval = 1.99, 2.72). A negative linear correlation between the average concentration of MDA during the observation period and the average concentration of TAC for all the animals was observed (p < 0.001, R ² = −0.7868). Moreover, a negative linear correlation between the average concentration of MDA during the observation period and the average concentration of TAC in LDA group (p < 0.001, R ² = −0.809) and control group (p < 0.001, R ² = −0.762) was observed respectively (Figure 2). In the first sampling time for the control group, we noted a significant negative correlation between the average concentration of MDA and TAC (p < 0.001, R ² = −0.9657). Conversely, there was an absence of correlation between the average concentration of MDA and TAC in the LDA group during the same sampling time (p = 0.86, R ² = 0.080). In the second sampling time for the control group, we noted a significant negative correlation between the average concentration of MDA and TAC (p = 0.005, R ² = −0.899). Conversely, there was an absence of correlation between the average concentration of MDA and TAC in the LDA group during the same sampling time (p = 0.68, R ² = 0.19). In the third sampling time for the control group, there was no correlation between the average concentration of MDA and TAC (p = 0.899, R ² = 0.059). Furthermore, there was an absence of correlation between the average concentration of MDA and TAC in the LDA group during the same sampling time (p = 0.83, R ² = 0.10). In the fourth sampling time for the control group, we noted a significant negative correlation between the average concentration of MDA and TAC (p < 0.001, R ² = −0.966). Conversely, there was an absence of correlation between the average concentration of MDA and TAC in the LDA group during the same sampling time (p = 0.66, R ² = 0.20).

FIGURE 2.

Correlation between average malondialdehyde concentration and total antioxidant capacity (TAC) during the transitional period in Holstein cows in different experimental groups (LDA‐affected cows and control group) (p < 0.001 for LDA‐affected cows and control group). LDA, left displacement of the abomasum.

3.2. Metabolic Parameters

A significant effect of sampling time (p < 0.001) and treatment (p = 0.02) on the average concentration of urea was observed. The average concentration of urea was higher in cows with LDA (mean = 63.57 mg/dL, 95% confidence interval = 55.86, 71.28) compared to the healthy cows (mean = 49.85 mg/dL, 95% confidence interval = 42.15, 57.57).

A significant effect of sampling time (p < 0.001) and treatment (p = 0.02) on the average concentration of BUN was observed. The average concentration of BUN was higher in cows with LDA (mean = 29.70 mg/dL, 95% confidence interval = 26.10, 33.30) compared to the healthy cows (mean = 23.28 mg/dL, 95% confidence interval = 19.70, 26.90). A negative linear correlation was observed between the average concentration of BUN and TAC for all animals (p = 0.003, R ² = −0.38). Conversely, a positive linear correlation was found between the average concentration of BUN and MDA for all animals (p < 0.001, R ² = 0.47). In the LDA group, a negative linear correlation was noted between the average concentration of BUN and TAC (p = 0.002, R ² = −0.55), whereas no correlation was observed in the control group (p = 0.86, R ² = −0.053). Additionally, a positive linear correlation was found between the average concentration of BUN and MDA in the LDA group (p < 0.001, R ² = 0.60), but no correlation was detected in the control group (p = 0.78, R ² = −0.053).

A significant effect of sampling time (p < 0.001) and treatment (p < 0.001) on the average concentration of Cl was observed. Furthermore, the average concentration of Cl was (p < 0.001) affected by the interaction of sampling time and treatment. Average concentrations of Cl across different sampling times in two different treatments are presented in Table 6. The average concentration of Cl was lower in cows with LDA (mean = 76.58 mmol/L, 95% confidence interval = 73.73, 79.43) compared to the healthy cows (mean = 82.38 mmol/L, 95% confidence interval = 79.52, 85.23). A positive linear correlation was observed among the average concentrations of Cl and TAC for all animals (p = 0.002, R ² = 0.59). Conversely, a negative linear correlation was found between the average concentration of Cl and MDA for all animals (p < 0.001, R ² = −0.66). In the LDA group, a positive linear correlation was noted among the average concentrations of Cl and TAC (p < 0.001, R ² = 0.64), whereas no correlation was observed in the control group (p = 0.42, R ² = 0.38). Additionally, a negative linear correlation was found between the average concentration of Cl and MDA in the LDA group (p < 0.001, R ² = −0.66). Furthermore, a negative correlation was detected in the control group between the average concentration of Cl and MDA (p = 0.008, R ² = −0.49).

A significant effect of sampling time (p < 0.001) and interaction of treatment and sampling time (p = 0.001) on the average concentration of Ca was observed. Average concentrations of Ca across different sampling times in two different treatments are presented in Table 6. A positive linear correlation was observed between the average concentration of Ca and TAC for all animals (p = 0.002, R ² = 0.47). Conversely, a negative linear correlation was found between the average concentration of Ca and MDA for all animals (p < 0.001, R ² = −0.50). In the LDA group, a positive linear correlation was noted between the average concentration of Ca and TAC (p < 0.001, R ² = 0.62), whereas no correlation was observed in the control group (p = 0.12, R ² = 0.30). Additionally, a negative linear correlation was found between the average concentration of Ca and MDA in the LDA group (p = 0.001, R ² = −0.57). Furthermore, a negative correlation was detected in the control group between the average concentration of Ca and MDA (p = 0.003, R ² = −0.54).

A significant effect of sampling time (p < 0.001) and interaction of treatment and sampling time (p < 0.001) on the average concentration of P was observed. Average concentrations of P in different sampling times are presented in Table 5. A positive linear correlation was observed between the average concentration of P and TAC for all animals (p = 0.02, R ² = 0.30). Conversely, a negative linear correlation was found between the average concentration of P and MDA for all animals (p = 0.006, R ² = −0.36). In the LDA group, a positive linear correlation was noted between the average concentration of P and TAC (p < 0.001, R ² = 0.68), whereas no correlation was observed in the control group (p = 0.68, R ² = −0.19). Additionally, a negative linear correlation was found between the average concentration of P and MDA in the LDA group (p = 0.006, R ² = −0.70). However, there was no correlation in the control group between the average concentration of P and MDA (p = 0.65, R ² = −0.09).

TABLE 5.

Average serum concentration of measured parameters in four different sampling times and two treatments, including control and left displacement of the abomasum (LDA) cows.

	Treatments (mean ± SE)			Sampling times (mean ± SE)					T × St
Variable	Control	LDA	p	1	2	3	4	p	p
Urea (mg/dL)	49.86 ± 3.53	63.57 ± 3.53	0.02	42.07 ± 3.56a,b	51.86 ± 3.56b	58.57 ± 3.56c	74.36 ± 3.56d	<0.001	0.12
BUN (mg/dL)	23.28 ± 1.65	29.71 ± 1.65	0.02	19.66 ± 1.70a	24.23 ± 1.70b,a,c	27.40 ± 1.70c	34.75 ± 1.70d	<0.001	0.12
GGT (U/L)	22.04 ± 1.32	28.20 ± 2.16	0.02	19.07 ± 1.28a	25.15 ± 2.22b,c,d	31.30 ± 3.44c	26.67 ± 2.5d,c	0.001	0.004
TB (g/dL)	28.57 ± 1.70	27.02 ± 1.52	0.43	44.32 ± 4.68a	27.14 ± 1.76b,c	24.52 ± 1.43c,d	22.18 ± 1.73d	<0.001	0.23
AL (g/dL)	26.41 ± 0.71	28.08 ± 0.71	0.12	26.25 ± 0.72A	27.52 ± 0.72	26.76 ± 0.72	28.45 ± 0.72B	0.07	<0.001
TP (g/dL)	7.33 ± 0.10	7.20 ± 0.10	0.36	7.32 ± 0.13	7.33 ± 0.13	7.03 ± 0.13	7.38 ± 0.13	0.18	0.24
Gl (g/dL)	4.69 ± 0.14	4.39 ± 0.14	0.16	4.70 ± 0.14	4.58 ± 0.14	4.35 ± 0.14	4.53 ± 0.14	0.26	0.93
Al/Gl	5.71 ± 0.36	6.57 ± 0.36	0.12	5.69 ± 0.33	6.15 ± 0.33	6.35 ± 0.33	6.36 ± 0.33	0.21	0.06
Glu (mg/dL)	72.54 ± 6.20	95.93 ± 6.20	0.02	61.18 ± 7.28a	81.46 ± 7.28b,a	108.64 ± 7.28c	85.64 ± 7.28	<0.001	<0.001
Chol (mg/dL)	3.06 ± 0.16	2.77 ± 0.16	0.03	2.59 ± 0.18	2.96 ± 0.18	2.79 ± 0.18	3.31 ± 0.18	0.22	0.35
TG (mg/dL)	18.69 ± 1.30	28.27 ± 1.30	<0.001	18.51 ± 1.66a	26.37 ± 1.66b	24.11 ± 1.66	24.94 ± 1.66c,b	0.008	<0.001
NEFA (mmol/L)	0.52 ± 0.01	0.53 ± 0.01	0.68	0.52 ± 0.01	0.49 ± 0.01A	0.53 ± 0.014	0.54 ± 0.01B	0.07	<0.001
BHBA (mmol/L)	0.75 ± 0.03	1.10 ± 0.03	<0.001	0.75 ± 0.04a,b	0.85 ± 0.04b	1.08 ± 0.04c	1.03 ± 0.04d,c	<0.001	<0.001
AST (U/L)	77.91 ± 5.66	107.38 ± 5.66	<0.001	51.75 ± 8.01a	91.20 ± 8.01b	133.94 ± 8.01c	93.68 ± 8.01d,b	<0.001	<0.001
SDH (U/L)	10.47 ± 0.41	12.16 ± 0.41	0.01	10.81 ± 0.55	11.74 ± 0.55	10.74 ± 0.55	11.96 ± 0.55	0.27	0.27
SAA (µg/mL)	2.61 ± 0.26	4.93 ± 0.26	<0.001	3.14 ± 0.29a	3.95 ± 0.29	3.70 ± 0.29	4.30 ± 0.28b	0.02	<0.001
Cl (mmol/L)	82.38 ± 1.31	76.58 ± 1.31	<0.001	88.31 ± 1.60a	77.09 ± 1.60b	75.83 ± 1.60c,b	76.68 ± 1.60d,b,c	<0.001	<0.001
Ca (mg/dL)	7.40 ± 0.13	7.28 ± 0.13	0.51	8.10 ± 0.15a	6.80 ± 0.15b	6.86 ± 0.15c,b	7.59 ± 0.15d	<0.001	0.001
P (mg/dL)	5.30 ± 0.12	5.29 ± 0.12	0.95	6.11 ± 0.24a	5.41 ± 0.24	4.78 ± 0.24b	4.88 ± 0.24c,b	<0.001	<0.001
Na (mg/dL)	121.6 ± 1.05	104.21 ± 1.05	<0.001	119.9 ± 1.44a	110.9 ± 1.44b	108.6 ± 1.44c,b	112.2 ± 1.44d,b,c	<0.001	<0.001
K (mg/dL)	4.3 ± 1.44	3.35 ± 1.44	<0.001	3.40 ± 0.18	3.71 ± 0.18	3.65 ± 0.18	3.40 ± 0.18	0.41	0.002
TAC (mmol/L)	0.93 ± 0.04	0.80 ± 0.04	0.03	1.07 ± 0.04a	0.87 ± 0.04b	0.72 ± 0.04c,b	0.80 ± 0.04d,b,c	<0.001	<0.001
MDA (nmol/mL)	2.36 ± 0.17	3.76 ± 0.17	<0.001	1.82 ± 0.18a	2.96 ± 0.18b	4.04 ± 0.18c	3.42 ± 0.18d,b	<0.001	<0.001

Note: Interaction of treatment and sampling time. Within rows, values with different superscripts (a,b) differ (p ≤ 0.05). Within rows, values with different superscripts (A,B) differ (0.05 < p ≤ 0.10).

Abbreviations: AL/Gl, albumin/globulin ratio; AL, albumin; AST, aspartate aminotransferase; BHBA, β‐hydroxybutyric acid; Ca, calcium; Chol, cholesterol; Cl, chloride; Gl, globulin; Glu, glucose; K, potassium; MDA, malondialdehyde; Na, sodium; NEFA, non‐esterified fatty acids; P, phosphorous; SAA, serum amyloid A; SDH, succinate dehydrogenase; TAC, total antioxidant capacity; TB, total bilirubin; TG, triglyceride; TP, total protein.

A significant effect of sampling time (p < 0.001) and treatment (p < 0.001) on the average concentration of Na was observed. Furthermore, the average concentration of Na was (p < 0.001) affected by the interaction of sampling time and treatment. Average concentrations of Na across different sampling times in two different treatments are presented in Table 6. The average concentration of Na was lower in cows with LDA (mean = 104.21 mg/dL, 95% confidence interval = 101.94, 106.49) compared to the healthy cows (mean = 121.60 mg/dL, 95% confidence interval = 119.29, 123.85). Average concentrations of Na in different sampling times are presented in Table 5. A positive linear correlation was observed between the average concentration of Na and TAC for all animals (p < 0.001, R ² = 0.61). Conversely, a negative linear correlation was found between the average concentration of Na and MDA for all animals (p < 0.001, R ² = −0.78). In the LDA group, a positive linear correlation was noted between the average concentration of Na and TAC (p < 0.001, R ² = 0.81), whereas no correlation was observed in the control group (p = 0.07, R ² = 0.34). Additionally, a negative linear correlation was found between the average concentration of Na and MDA in the LDA group (p < 0.001, R ² = −0.78). Furthermore, a negative correlation was detected in the control group between average concentration of Na and MDA (p = 0.01, R ² = −0.47).

A significant effect of treatment (p < 0.001) and interaction of treatment and sampling time (p = 0.002) on the average concentration of K was observed. Average concentrations of K across different sampling times in two different treatments are presented in Table 6. The average concentration of K was lower in cows with LDA (mean = 3.35 mg/dL, 95% confidence interval = 3.07, 3.64) compared to the healthy cows (mean = 4.30 mg/dL, 95% confidence interval = 4.02, 4.59). The correlation analysis results for the average concentration of K and oxidative stress biomarkers and variations in the average concentration of K across different sampling times are detailed in Tables 4 and 5, respectively.

TABLE 4.

Pearson's correlation of the selected metabolic factors and liver enzymes, and oxidative stress biomarkers.

	MDA						TAC
	All cows		LDA cows		Control cows		All cows		LDA cows		Control cows
Variables	r a	p value b	r	p value	r	p value	r	p value	r	p value	r	p value
MDA c	1.000		1.000		1.000		−0.7868	p < 0.001	−0.809	p < 0.001	−0.762	p < 0.001
BUN d	0.47	p < 0.001	0.60	p < 0.001	−0.053	p = 0.78	−0.38	p = 0.003	−0.55	p = 0.002	−0.035	p = 0.86
Glucose	0.60	p < 0.001	0.79	p < 0.001	0.39	p = 0.04	−0.56	p < 0.001	−0.71	p = 0.002	−0.16	p = 0.41
TB e	0.42	p = 0.001	0.64	p < 0.001	0.08	p = 0.65	−0.46	p < 0.001	−0.84	p = 0.002	−0.05	p = 0.78
AST f	0.56	p = 0.001	0.66	p < 0.001	0.24	p = 0.20	−0.47	p < 0.001	−0.69	p = 0.002	−0.12	p = 0.55
SAA g	0.53	p < 0.001	0.70	p < 0.001	−0.17	p = 0.38	−0.35	p = 0.008	−0.64	p = 0.002	0.12	p = 0.53
BHBA h	0.66	p < 0.001	0.69	p < 0.001	0.13	p = 0.51	−0.59	p < 0.001	−0.65	p < 0.001	−0.22	p = 0.26
Triglyceride	0.54	p < 0.001	0.66	p < 0.001	0.08	p = 0.78	−0.42	p < 0.001	−0.69	p < 0.001	0.15	p = 0.43
Chloride	−0.66	p < 0.001	−0.66	p < 0.001	−0.49	p = 0.008	0.59	p = 0.002	0.64	p < 0.001	0.38	p = 0.42
Sodium	−0.78	p < 0.001	−0.78	p < 0.001	−0.47	p = 0.01	0.61	p < 0.001	0.81	p < 0.001	0.34	p = 0.07
Potassium	−0.41	p = 0.03	−0.54	p < 0.001	−0.30	p = 0.12	0.38	p = 0.03	0.41	p = 0.004	0.14	p = 0.48
Calcium	−0.50	p < 0.001	−0.57	p = 0.001	−0.54	p = 0.003	0.47	p = 0.002	0.62	p < 0.001	0.30	p = 0.12
Phosphorous	−0.36	p = 0.006	−0.70	p = 0.006	−0.09	p = 0.65	0.30	p = 0.02	0.68	p < 0.001	−0.19	p = 0.68

Abbreviations: AST, aspartate aminotransferase; BUN, blood urea nitrogen; LDA, left displacement of the abomasum; MDA, malondialdehyde; SAA, serum amyloid A; TAC, total antioxidant capacity.

^a r, for the correlation coefficient.

^b p value for the correlation of the two variables.

^c Malondialdehyde.

^d Blood urea nitrogen.

^e Total bilirubin.

^f Aspartate amino transferase.

^g Serum amyloid A.

^h β‐Hydroxybutyric acid.

A significant effect of sampling time (p < 0.001) and treatment (p = 0.02) on the average concentration of glucose was observed. Furthermore, the average concentration of glucose was (p < 0.001) affected by the interaction of sampling time and treatment, which is presented in Table 6. The average concentration of glucose was higher in cows with LDA (mean = 95.93 g/dL, 95% confidence interval = 82.42, 109.43) compared to the healthy cows (mean = 72.54 g/dL, 95% confidence interval = 59.03, 86.04). Average concentrations of glucose in different sampling times are presented in Table 5. A negative linear correlation was observed between the average concentration of glucose and TAC for all animals (p < 0.001, R ² = −0.56). Conversely, a positive linear correlation was found between the average concentration of glucose and MDA for all animals (p < 0.001, R ² = 0.60). In the LDA group, a negative linear correlation was noted between the average concentration of glucose and TAC (p = 0.002, R ² = −0.71), whereas no correlation was observed in the control group (p = 0.41, R ² = −0.16). Additionally, a positive linear correlation was found between the average concentration of glucose and MDA in the LDA group (p < 0.001, R ² = 0.79). Moreover, a positive linear correlation was detected in the control group between glucose and MDA (p = 0.04, R ² = 0.39).

A significant effect of sampling time (p = 0.008) and treatment (p < 0.001) on the average concentration of TAG was observed. Furthermore, the average concentration of TAG was (p < 0.001) affected by the interaction of sampling time and treatment, which is presented in Table 6. The average concentration of TAG was higher in cows with LDA (mean = 28.27 mg/dL, 95% confidence interval = 25.45, 31.09) compared to the healthy cows (mean = 18.69 mg/dL, 95% confidence interval = 15.87, 21.52). The correlation analysis results for the average concentration of TAG and oxidative stress biomarkers and variations in the average concentration of TAG across different sampling times are detailed in Tables 4 and 5, respectively.

Sampling time tends to (p = 0.07) affect the average concentration of NEFA. Furthermore, the interaction of sampling time and treatment affects (p < 0.001) the average concentration of NEFA. Average concentration of NEFA across different sampling times is presented in Table 5.

A significant effect of sampling time (p < 0.001) and treatment (p < 0.001) on the average concentration of BHBA was observed. Furthermore, the average concentration of BHBA was (p < 0.001) affected by the interaction of sampling time and treatment, which is presented in Table 6. The average concentration of BHBA was higher in cows with LDA (mean = 1.10 mmol/L, 95% confidence interval = 1.03, 1.18) compared to the healthy cows (mean = 0.75 mmol/L, 95% confidence interval = 0.68, 0.83). Average concentration of BHBA across different sampling times is presented in Table 5. Average concentrations of BHBA in different sampling times are presented in Table 5. A negative linear correlation was observed between the average concentration of BHBA and TAC for all animals (p < 0.001, R ² = −0.59). Conversely, a positive linear correlation was found between the average concentration of BHBA and MDA for all animals (p < 0.001, R ² = 0.66). In the LDA group, a negative linear correlation was noted between the average concentration of BHBA and TAC (p = 0.002, R ² = −0.65), whereas no correlation was observed in the control group (p = 0.26, R ² = −0.22). Additionally, a positive linear correlation was found between the average concentration of BHBA and MDA in the LDA group (p < 0.001, R ² = 0.69). However, no correlation was detected in the control group between BHBA and MDA (p = 0.51, R ² = 0.13).

A significant effect of sampling time (p = 0.001) and treatment (p = 0.02) on the average concentration of GGT was observed. Moreover, the average concentration of GGT was (p = 0.004) affected by the interaction of sampling time and treatment, which is presented in Table 6. The average concentration of GGT was higher in cows with LDA (mean = 28.20 U/L, 95% confidence interval = 24.16, 33.85) compared to the healthy cows (mean = 22.04 U/L, 95% confidence interval = 19.50, 25.36).

A significant effect of sampling time (p < 0.001) on the average concentration of total bilirubin was observed. The correlation analysis results for the average concentration of total bilirubin and oxidative stress biomarkers and variations in the average concentration of total bilirubin across different sampling times are detailed in Tables 4 and 5, respectively.

A significant effect of sampling time (p < 0.001) and treatment (p < 0.001) on the average concentration of AST was observed. Furthermore, the average concentration of AST was (p < 0.001) affected by the interaction of sampling time and treatment, which is presented in Table 6. The average concentration of AST was higher in cows with LDA (mean = 107.38 U/L, 95% confidence interval = 95.99, 118.76) compared to the healthy cows (mean = 77.91 U/L, 95% confidence interval = 66.52, 89.29). Average concentration of AST across different sampling times is presented in Table 5. A negative linear correlation was observed between the average concentration of AST and TAC for all animals (p < 0.001, R ² = −0.47). Conversely, a positive linear correlation was found between the average concentration of AST and MDA for all animals (p = 0.001, R ² = 0.56). In the LDA group, a negative linear correlation was noted between the average concentration of AST and TAC (p = 0.002, R ² = −0.69), whereas no correlation was observed in the control group (p = 0.55, R ² = −0.12). Additionally, a positive linear correlation was found between the average concentration of AST and MDA in the LDA group (p < 0.001, R ² = 0.66). However, no correlation was detected in the control group between AST and MDA (p = 0.20, R ² = 0.24).

A significant effect of treatment (p = 0.01) on the average concentration of SDH was observed. The average concentration of SDH was higher in cows with LDA (mean = 12.16 U/L, 95% confidence interval = 11.27, 13.05) compared to the healthy cows (mean = 10.47 U/L, 95% confidence interval = 9.57, 11.36).

A significant effect of sampling time (p = 0.02) and treatment (p < 0.001) on the average concentration of SSA was observed. Furthermore, the average concentration of SSA was (p < 0.001) affected by the interaction of sampling time and treatment, which is presented in Table 6. The average concentration of SSA was higher in cows with LDA (mean = 4.93 µg/mL, 95% confidence interval = 4.36, 5.49) compared to the healthy cows (mean = 2.61 µg/mL, 95% confidence interval = 2.04, 3.18). The correlation analysis results for the average concentration of SAA and oxidative stress biomarkers and variations in the average concentration of SAA across different sampling times are detailed in Tables 4 and 5, respectively.

The average concentration of other measured parameters across various treatments and sampling times, as well as the breakdown for each sampling time into the healthy group and LDA group, is detailed in Tables 5 and 6, along with their respective standard errors of the mean.

4. Discussion

The association between alterations in oxidative stress biomarkers and other relevant blood parameters associated with LDA remains relatively unexplored. Furthermore, inconsistent findings have been reported regarding the oxidant and antioxidant levels in cows with LDA. Although studies by Maden et al. (2012) and Durgut et al. (2016) did not observe significant differences between healthy cows and those with LDA, Hasanpour, Saranjam, and Amuoghli Tabrizi (2011), Mamak et al. (2013) and Aly et al. (2016) reported elevated oxidation parameters and reduced antioxidant capacity in cows with LDA. These inconstant findings may imply the involvement of multiple factors in determining the oxidative status of cows with LDA. The observed substantial rise in MDA levels and a notable decline in TAC in cows with LDA compared to their healthy counterparts in this study provided compelling evidence that LDA is linked to alterations in the animals’ oxidative status (Hasanpour, Saranjam, and Amuoghli Tabrizi 2011; Qu et al. 2013; Durgut et al. 2016). Our findings align with Maden et al. (2012), who also reported an increase in MDA concentration in cows with LDA, attributing it to tissue damage caused by hypoxia. Current results agree with previous findings, providing support for the association between LDA and oxidative stress.

Lipid peroxidation, a chain reaction involving the oxidative breakdown of lipids, leads to the production of MDA as the final byproduct (Sharma et al. 2011). In our study, MDA concentrations were approximately twofold higher in the third sampling compared to the first sampling, indicating an increased occurrence of lipid peroxidation during the transitional period, which aligns with previous findings (Saleh, Salam, and Mileegy 2007; Sharma et al. 2011). Furthermore, TAC concentration exhibited lower values in cows in the third sampling time compared to the first sampling time. Additionally, our observation of a declining trend in TAC concentration as parturition approached is consistent with the findings reported by Brezezinska et al. (1994), Chawla and Kaur (2004), Chatterjee, Kaur, and Panda (2003) and Khatti et al. (2017). Current findings suggest that all animals undergo oxidative stress during the transitional period, regardless of their LDA status. However, cows with LDA experience heightened oxidative stress in addition to physiological oxidative stress. LDA contributes to an increased oxidative stress burden. Considering both factors is crucial in assessing the oxidative status of dairy cows and managing LDA. Additionally, a strong negative linear correlation between the average MDA concentration and the average TAC concentration was observed during the transitional period. This correlation was more pronounced in cows with LDA compared to healthy cows. It is noteworthy to mention that it is evident that there is a discernible alteration in the trends of indicator values (MDA and TAC) among LDA cows before calving, deviating from the patterns observed in healthy counterparts. This shift in trends may indicate distinct metabolic processes in these animals, potentially initiated before calving, as metabolic diseases can manifest before parturition, which might be an explanation for the inconsistent results presented by the previous studies. Additionally, the limited sample size of LDA cows in the current study and previous experiments presents a challenge, as it could be influenced by unknown individual differences among the animals, potentially impacting the observed indicator values.

Previous studies linked NEB, indicated by increased NEFA and BHBA, to LDA risk (LeBlanc, Leslie, and Duffield 2005). Consistent with these studies, cows with LDA had higher BHBA concentrations, and a slight increase in the average NEFA concentration was observed. During the first, third and fourth sampling times, NEFA and BHBA concentrations were higher in the LDA group compared to healthy cows. TAG and cholesterol levels differed among treatment groups, with higher concentrations in cows with LDA. In the LDA group, TAG levels progressively increased over time, reflecting ongoing fat mobilization. These findings suggest increased lipolysis and fat mobilization in cows with LDA, leading to elevated circulating TAG levels. In the current study, significant correlations between BHBA and TAG concentrations with MDA concentration were observed. Additionally, a negative correlation was observed between BHBA and TAG concentrations with TAC concentration in both healthy cows and those with LDA during the transitional period. These correlations were particularly notable in cows with LDA, emphasizing metabolic changes associated with oxidative status and the role of lipid metabolism in LDA development. Current findings agree with previous studies (Bernabucci et al. 2005; Shi et al. 2014), supporting the positive association among elevated BHBA levels, increased oxidation (higher MDA levels) and decreased antioxidant markers. This finding further supports the notion that higher BHBA levels during the transitional period, including in cows with LDA, contribute to heightened oxidative stress and compromised antioxidant defence and might be the predisposing factor for LDA.

LDA‐diagnosed cows showed elevated GGT, AST and SDH concentrations compared to healthy cows, supporting prior research indicating higher liver enzyme levels in cows with LDA (Komatsu et al. 2002; Klevenhusen et al. 2015). Furthermore, GGT and AST concentrations displayed an upward trend during the transitional period. The higher concentrations of SDH and GGT in cows with LDA may indicate potential hepatocyte injury (Kaneko, Harvey, and Bruss 1997; El‐Kabbani, Darmanian, and Chung 2004; Kalaitzakis et al. 2007). Similarly, the elevated AST concentration in cows with LDA could be associated with increased liver fat content, consistent with previous studies (Komatsu et al. 2002). During the transitional period, we observed a positive correlation between total bilirubin and AST concentration with MDA concentration. Additionally, a negative correlation between total bilirubin and AST with TAC was observed. Importantly, these correlations were stronger in cows with LDA. Positive correlation of AST and MDA has been reported in ketonic cows previously (Li et al. 2016), suggesting a link between hepatic dysfunction and alteration in oxidation/antioxidant balance. These findings highlight the interplay among liver health, oxidative stress and antioxidant capacity in the context of LDA.

In this study, cows with LDA exhibited higher glucose concentrations than healthy cows. This aligns with previous findings indicating that cows with prolonged NEFA concentration increases during early lactation may have impaired glucose clearance (Holtenius et al. 2003). NEFA has been shown to have an impact on glucose tolerance and insulin responsiveness in cows (Pires, Souza, and Grummer 2007; Sordilo and Mavangira 2014). High NEFA concentrations in cows with LDA can contribute to elevated glucose levels. These findings emphasize the intricate relationship among NEFA metabolism, glucose regulation and LDA development in dairy cows. Insulin resistance commonly occurs in cows with LDA. Mild increases in plasma insulin concentrations, independent of ketosis, are frequently seen in cows with LDA, along with hyperglycaemia. This condition is often accompanied by reduced milk production, as observed in our study (Constable et al. 2016). Evidence suggests a potential correlation between increased glucose levels and oxidative stress. Elevated glucose levels are often linked to conditions like insulin resistance. ROS have been shown to hinder insulin production. The lower expression of antioxidants in beta cells, compared to other tissues, may contribute to oxidative stress and the decline of beta cells in Type 2 diabetes, as proposed by Kaneto et al. (2007). Abuelo et al. (2016) found that cows at a higher risk of oxidative stress, marked by increased ROS production and/or reduced antioxidant capacity, tend to have lower insulin sensitivity in peripheral tissues. During the transitional period, a positive correlation between glucose concentration and MDA levels, and the negative correlation between glucose concentration and TAC, was seen. These correlations were more pronounced in cows with LDA, suggesting a potential association between elevated glucose levels and oxidative stress. This study is the first to report the correlation between glucose and oxidative stress biomarkers in cows with LDA. However, the mechanisms involved are complex and multifactorial, requiring further research to elucidate the molecular basis of this relationship.

The higher concentration of BUN observed in our study is consistent with previous findings (de Cardoso et al. 2008; Song et al. 2020). An increase in BUN as a reliable indicator of renal function as it reflects the final product of protein metabolism is typically associated with a decline in glomerular filtration rate, which occurs when the renal function is compromised (Staric et al. 2010). In the context of our study, it is possible that the observed increase in BUN in cows with LDA may be a result of haemoconcentration and dehydration, which can be consequences of LDA. During the transitional period, we observed a positive correlation between BUN concentration and MDA levels. Additionally, there was a negative correlation between BUN and TAC. Emerging evidence indicates a potential link between kidney damage occurring in a hyperlipaemia state and an elevation in oxidative stress (Scheuer et al. 2000). Hence, it can be hypothesized that there may be a concurrent kidney injury characterized by an elevation in BUN levels. This kidney injury may coincide with the presence of oxidative stress in the cows during the transitional period, especially in cows with LDA, as observed in the present results.

Increase in acute‐phase proteins such as SAA has been reported as a consequence of endotoxaemia (Chalmeh et al. 2016; Shahraki et al. 2016). However, in transition period, dairy cows may exhibit a noticeable systemic inflammatory response, even in the absence of apparent microbial infections (Bertoni et al. 2008). Cows with moderate‐to‐severe fatty livers had higher levels of the pro‐inflammatory cytokines (Ohtsuka et al. 2001). In this study, an increase in SAA during the transitional period indicated an elevated inflammatory response. Similar to the findings of Guzelbektes et al. (2010), cows with LDA exhibited approximately twofold higher SAA concentration compared to healthy cows, signifying an inflammatory condition. The existing inflammation during the transitional period may further enhance the elevated SAA levels in cows with LDA. A positive correlation between SAA and MDA levels and the negative correlation between SAA and TAC during this period, especially in cows with LDA, was observed. Evidence suggests a potential correlation between increased SAA levels and MDA in bovine respiratory disease in dairy calves (Joshi et al. 2018). Limited research directly explores the correlation between SAA and oxidative stress, particularly in cows with LDA, indicating a plausible association between elevated SAA levels and heightened oxidative stress that requires further investigation.

Mild metabolic alkalosis, hypochloraemia and hypokalaemia are observed in cows with LDA due to abomasal hypomotility, continued hydrochloric acid secretion and impaired flow into the duodenum (Constable et al. 2016). In this study, cows with LDA had lower Cl, K and Na concentrations compared to healthy cows, consistent with previous findings. During the transitional period, negative correlations were observed among Cl, Na and K levels and MDA concentration, and positive correlations were observed with TAC concentration. These correlations were more pronounced in cows with LDA. pH imbalances during the transitional period can impact a cow's oxidative/antioxidant capacity. Implementing a DCAD diet has been shown to decrease MDA concentration, indicating the influence of ion disruptions caused by abomasal displacement (Wu et al. 2013; Rajaeerad et al. 2021). Further investigation can provide insights into the influence of mineral imbalances on oxidative stress.

In the transitional period, a negative correlation was found between Ca and MDA. Conversely, a positive correlation existed between Ca and TAC. These associations were more pronounced in cows with LDA. Ca plays a crucial role in the regulation and activation of various antioxidant enzymes, including superoxide dismutase, catalase and glutathione peroxidase, enabling them to efficiently neutralize ROS and protect cells from oxidative damage (Nazıroğlu et al. 2012, 2011; Szlacheta et al. 2020). Other studies have documented that the addition of 25‐hydroxyvitamin D3 and oral calcium bolus as supplements can lead to a reduction in oxidative stress in dairy cows during the transitional period (Xu et al. 2021) and other species (Yang et al. 2019). Although the exact causal relationship between changes in calcium concentration and alterations in oxidative/antioxidant biomarkers cannot be conclusively determined in this study, the simultaneous occurrence of these changes, likely due to calving, is significant. Further research is needed to elucidate the direct association between calcium and oxidative biomarkers and their potential impacts on the health of transition cows.

Limited by the study design, although the influence of decreased dry matter intake on LDA has been proposed, individual food intake data were not obtainable. However, food consumption was monitored between two groups. Recent findings have highlighted the considerable variability in feeding‐related behaviours, including rumination, among animals (Mirzaei et al. 2023). Therefore, a limitation of this study was the unavailability of individual food intake data. Furthermore, caution must be exercised when generalizing the study's outcomes to other farms, given its single‐farm setting.

This study has underscored the alterations in various metabolic and oxidative stress biomarkers and their variability during the transitional period in both healthy cows and those affected by LDA following the conducted samplings. Our findings elucidate shifts in metabolic, oxidative and certain liver‐associated parameters, indicating the comprehensive impact on multiple organs and systems during this crucial period, particularly with a greater magnitude in cows affected by LDA. Nonetheless, various feed supplement strategies have been explored to mitigate the intensity of these changes in dairy cows (Mirzaei et al. 2022), which can be used for further investigations. Moreover, correlations between oxidative stress biomarkers and indicative blood parameters of LDA highlight the complex nature of the condition and its systemic disturbances. Positive associations between MDA and metabolic parameters like NEFA and BUN suggest synergistic effects and underlying pathophysiological processes. Meanwhile, the negative correlation with TAC indicates compromised antioxidant defence mechanisms in LDA cows. These findings enhance our understanding of LDA mechanisms and guide future research and management strategies in dairy cattle. Recognizing the link among oxidative stress, metabolic imbalances and liver dysfunction enables interventions to mitigate LDA's detrimental effects and enhance cow health and productivity during the transitional period. Further investigations are necessary to fully comprehend the role of oxidative stress in LDA pathogenesis in dairy cows.

5. Conclusion

In conclusion, our study might shed light on the intricate interplay among oxidative stress, metabolic parameters and liver enzymes during the transitional period in dairy cows. The findings highlight the distinct differences observed in LDA compared to healthy cows, indicating an elevated state of oxidative stress and compromised antioxidant defence in LDA cows. These alterations are accompanied by changes in metabolic parameters, such as NEFA, glucose and urea levels, as well as liver enzyme concentrations.

Author Contributions

Ahmadreza Mirzaei: conceptualization, data curation, formal analysis, methodology, software, writing–original draft, writing–review and editing. Ali Hajimohammadi: conceptualization, funding acquisition, investigation, project administration, supervision, writing–review and editing. Amirhossein Nasrian: investigation. Mohammad Nikzad: data curation, investigation. Abbas Rowshan‐Ghasrodashti: data curation, investigation; supervision. Saeed Nazifi: funding acquisition, validation. Aboutorab Tabatabaei Naeini: investigation, validation, visualization.

Ethics Statement

All the calves used in this experiment were handled in accordance with the technical regulations and the guidelines set out by the committee of animal ethics of Shiraz University, Iran. The protocols of the study were approved by the Ethics Committee of Shiraz University (IACUC no.: 4687/63). We additionally noted the recommendations of European Council Directive (86/609/EC) of 24 November 1986 regarding the standards of protecting animals used for experimental purposes.

Conflicts of Interest

The authors declare no conflicts of interest.

Peer Review

The peer review history for this article is available at https://publons.com/publon/10.1002/vms3.70142.

Data Availability Statement

Data availability was declared by the authors.