Association of postpartum uterine diseases with lying time and metabolic profiles of multiparous Holstein dairy cows in the transition period

The Veterinary Journal 263 (2020) 105533 Contents lists available at ScienceDirect The Veterinary Journal journal homepage: www.elsevier.com/locate/tvjl Association of postpartum uterine diseases with lying time and metabolic profiles of multiparous Holstein dairy cows in the transition period L. Cattaneo, V. Lopreiato, E. Trevisi* , A. Minuti Department of Animal Sciences, Food and Nutrition, Faculty of Agriculture, Food and Environmental Science, Università Cattolica del Sacro Cuore, 29122 Piacenza, Italy A R T I C L E I N F O Keywords: Lying behaviour Metritis Pedometer Retained placenta Transition period A B S T R A C T The objective of this study was to assess how uterine disorders alter the lying behaviour and plasma biomarkers in dairy cows. 34 multiparous cows were retrospectively classified into three groups according to the first uterine disorder that cows were diagnosed with: retained placenta (RP), metritis (MET), or healthy (H; cows without any clinical disease). Lying time (LT) and duration of lying bouts (LB) were monitored between 6 weeks prior to and 8 weeks after calving via the AfiAct II pedometer. Blood samples were collected routinely between 14 days before and 28 days after calving. Data was analysed using Proc MIXED of SAS ver. 9.4. Regardless of grouping, both LT and LB were longer (P < 0.01) in the prepartum period (774  16.6 min/ day and 89.9  2.1 min/bout) than in the first 28 days after calving (DFC; 653  16.7 min/day and 63.7  2.1 min/bout). Cows with RP had longer LT than healthy cows during the last 3 weeks before calving (837  30.9 vs. 735  27.1 min/day; P < 0.05). LT in cows with MET and healthy cows were not significantly different. The LB was similar among groups, averaging 76.1  3.4 min/bout in healthy cows, 73.2  3.8 min/bout in cows with RP, and 75.2  3.7 min/bout in cows with MET (P > 0.05). Compared with healthy cows, cows with RP laid down longer and stood up for shorter times (P < 0.05), particularly before calving. In addition, cows with RP had increased mobilization of body stores and more pronounced inflammatory status, as demonstrated by plasma haptoglobin (P = 0.04) and albumin (P < 0.01) concentrations. Our data suggest that automatic monitoring of lying behaviour could help identify cows at increased risk of developing certain disorders, such as RP. © 2020 Elsevier Ltd. All rights reserved. Introduction Lying behaviour is one of the main activities of dairy cows (Grant, 2007) and has priority even over feeding and social behaviour (Munksgaard et al., 2005). Daily lying time is positively correlated with healthy milk production, lactation stage (Bewley et al., 2010), parity, and body condition score (BCS; Westin et al., 2016), while deprivation of lying causes stress and could impair productivity and welfare (Cooper et al., 2007). Previous investigations reported that heifers have an inelastic demand of 12 h of lying time per day (Jensen et al., 2005). Although no specific requirements have been defined for dairy cows in commercial dairy farms, an average lying time of about 12 h per day was reported for cows housed in a free-stall barn (Gomez and Cook, * Corresponding author. E-mail address: erminio.trevisi@unicatt.it (E. Trevisi). http://dx.doi.org/10.1016/j.tvjl.2020.105533 1090-0233/© 2020 Elsevier Ltd. All rights reserved. 2010) and 8.5 h per day for grazing cows (Sepúlveda-Varas et al., 2014). Optimal lying behaviour can be prevented by numerous factors. Lying behaviour can be influenced by herd and feeding management (DeVries and von Keyserlingk, 2005), stocking density (Fregonesi et al., 2007; Maselyne et al., 2017), grouping strategies (Jensen and Proudfoot, 2017), bedding comfort, design of the freestall barn (Cook, 2019), health status (Sepúlveda-Varas et al., 2014), heat stress (Hillman et al., 2005; Cook et al., 2007), and seasonal changes (Steensels et al., 2012). Moreover, important differences in lying behaviour in the transition period between primiparous and multiparous cows have been reported (Neave et al., 2017). It is well-known that some diseases and disorders affect daily lying time; therefore, monitoring cow behaviour could be useful for predicting health problems (Weary et al., 2009). Lameness is associated with increased lying time and reduced frequency of lying bouts (Solano et al., 2016). Subclinical ketosis is related to longer lying time in multiparous cows (Kaufman et al., 2016). 2 L. Cattaneo et al. / The Veterinary Journal 263 (2020) 105533 Mastitis could reduce lying time, mainly due to discomfort associated with udder swelling (Siivonen et al., 2011). Metritis (MET) has also been linked to an increase in lying time during the week after diagnosis in primiparous cows, but not in multiparous cows (Barragan et al., 2018). Cows diagnosed with this disease have been reported to have reduced lying time and lying bouts during the 2 weeks before diagnosis (Neave et al., 2018). With the development of precision livestock farming (Berckmans, 2014; Halachmi et al., 2019), sensors are now employed in fields to continuously record activity and lying behaviour (number of steps; lying time; frequency and length of lying bouts). Sensors such as pedometers, collars, and ear tags are used mainly for heat detection, but also for daily monitoring of health and behaviour status of individual cattle (Calamari et al., 2014; Kaufman et al., 2016; Lopreiato et al., 2018, 2020). Uterine disorders such as retained placenta (RP) and MET are multifactorial and can impair milk production and fertility, resulting in economic losses (Laven and Peters, 1996; Giuliodori et al., 2013; Moretti et al., 2015). Early identification of animals at increased risk of developing these disorders could allow for early therapeutic intervention, which could ameliorate negative sequelae. Hence, the primary objective of the current study was to investigate associations between RP and MET and lying behaviour during the transition period in high-yielding Holstein dairy cows. A second objective was to assess metabolic and inflammatory biomarkers in plasma to evaluate responses to RP and MET and their potential utility as indicators of these disorders during the transition period. Health status Health status was evaluated by the same veterinarian during a weekly reproductive visit and was monitored daily by the herd staff. Abnormal behaviour and any concerns were reported to the veterinarian for subsequent examination. Identification of diseases and date of initial detection were recorded as they occurred. RP was diagnosed when the fetal membranes had not been completely expelled 12 h after calving (Beagley et al., 2010). Uterine discharge was evaluated twice a week after the morning feeding beginning at 5 days after calving. Before manual examination, a diluted iodine solution was used to clean the vulva and remove any faecal material. MET was diagnosed based on the classification proposed by Urton et al. (2005) when mucopurulent and foul-smelling discharge was detected (score = 2). Cows diagnosed with MET received an IM antibiotic treatment (Cloxalene Plus, FATRO S.p.A.) for 3 consecutive days starting on the day of detection. Signs of other disorders (e.g., ketosis, displacement of the abomasum, foot disorders, and mastitis) were noted if present. Animal grouping A total of 47 cows were involved in this study. Since the main aim was to investigate post-partum uterine disease, 12 cows diagnosed with other disorders were excluded from the study: six cows had mastitis, four cows had abomasal displacement, and two cows had lameness. One cow was removed from the analysis due to pedometer failure. The remaining 34 cows were retrospectively divided into three groups, according to the first clinical uterine disease diagnosed in the first 28 days after calving. We focused on the main uterine diseases typically occurring in the early postpartum period, i.e., RP and MET. If multiple uterine disorders were diagnosed, cows were categorized by the first diagnosed disorder because it was assumed that other disorders developed consequentially (Peeler et al., 1994). Thus, the three resulting groups were as follows: RP (n = 10 cows were diagnosed with retained placenta as the first uterine disease after calving); MET (n = 11 cows were diagnosed with MET as the first disease after calving); and healthy cows (H; n = 13 cows without any clinical diseases diagnosed in the first 28 DFC). Blood sampling Materials and methods Animal management The study was performed on a commercial dairy farm in the Po Valley in Italy from October 2018 and April 2019, in accordance with Italian laws on animal experimentation and ethics and authorisation by the Italian Health Ministry (Approval number, 484/2018-PR; Protocol number 7D5FE.5; Approval date, 26 February 2018). A total of 47 multiparous Holstein cows with a parity range of 2–6 (3.0  1.1, mean  standard deviation) were examined. They were housed in a free-stall barn within cubicles during the entire length of the trial (98 days). Before calving, cows were housed in a closed group pen. Immediately after calving, cows were moved to a postpartum pen; they were moved to the lactation pen after 7 days. The stocking density was kept at 100% for the entire trial period. Cows were milked twice daily at 3.00 am and 3.00 pm. During the dry period and lactation, cows were fed a total mixed ration delivered once daily in the morning. The chemical composition of the diets is reported in Table 1. Blood samples were collected at –14, –3, 1, 3, 7, and 28 days from calving (DFC) using jugular venipuncture into evacuated heparinized collection tubes (BD Vacutainer) before the morning feeding (7.30 am  30 min). Blood samples were immediately cooled in a water and ice bath. After collection, blood samples were centrifuged at 1900  g for 16 min at 4  C. Plasma biomarkers were analysed at 37  C by an automated clinical analyser (ILAB 650, Instrumentation Laboratory). The analysed metabolites, as described by Lopreiato et al. (2019), included glucose, cholesterol, nonesterified fatty acids (NEFA), β-hydroxybutyrate (BHB), urea, creatinine, calcium, magnesium, phosphorus, sodium, potassium, chlorine, zinc, haptoglobin, ceruloplasmin, globulin, total protein, aspartate aminotransferase– glutamate oxaloacetate transaminase (AST-GOT), gamma-glutamyl transferase (GGT), alkaline phosphatase, bilirubin, albumin, paraoxonase, myeloperoxidase, thiol groups, reactive oxygen metabolites (ROM), ferric reducing antioxidant power (FRAP), thiol groups (SHp), retinol, tocopherol, and β-carotene. Details of analytical procedures adopted in blood analysis are reported in Table S1 (Supplementary data). BCS, rectal temperature and milk yield Lying behaviour Lying behaviour was monitored between 6 weeks prior to and 8 weeks after calving with the AfiTag II pedometer (SAE Afikim), which has been validated for automatic measuring of lying behaviour (Henriksen and Munksgaard, 2019). This sensor, attached to the hind limb of each cow, registered activity and lying time continuously and the duration of each bout. BCS was assessed by the same operator after each blood sampling, as described by Mezzetti et al. (2019). In the period closest to calving, rectal temperature was measured with a digital thermometer simultaneously with blood sampling (–3, 1, 3, and 7 DFC). Milk yield was recorded daily from 1 to 28 DFC. Daily values for lying time, duration of lying bouts, and milk yield were expressed as average weekly data. Statistical analysis Table 1 Chemical composition of diets fed to dry and lactating cows. Dry matter (kg) Nutrients, % DM Starch Crude protein Ether extract Neutral detergent fiber Ash Energy calculated a Digestible energy, Mcal/kg of DM Net energy for lactation, Mcal/kg of DM DM, Dry matter. National Research Council (2001). Prepartum Lactation 12.6 23.5 5.3 11.9 2.2 51.1 9.3 26.6 16.4 4.9 31.0 6.5 2.6 1.3 2.91 1.58 Data in tables are presented as least squares means and standard error of the mean (LSM  SEM). Before statistical analysis, normality of the data was verified by calculating the kurtosis and asymmetry indices (Shapiro test, SAS Institute, Release 8.0). Parameters that were not normally distributed were subjected to logarithmic transformation (i.e., glucose, AST-GOT, GGT, bilirubin, alkaline phosphatase, NEFA, BHB, myeloperoxidase, FRAP, and tocopherol concentrations). After analysis, residuals were plotted to assess model assumptions of normality and homoscedasticity. Data for lying, blood parameters, BCS, rectal temperature, and milk yield were submitted to ANOVA and analysed by repeated measures in the MIXED procedure of SAS (SAS Institute, Version 9.4). The covariance structure (compound symmetry, autoregressive order, and spatial power) with the lowest AICC (Littell et al., 1998) was included in the MIXED model. Health status (HS; RP, MET, and H groups), time (–14, –3, 1, 3, 7, 28 DFC or weeks from parturition in the case of lying analysis and milk yield), and their interactions were used as fixed effects; cows were nested within the disorder group as the random effect. Data were considered significant at P  0.05, using the PDIFF statement in SAS. In addition, when a significant L. Cattaneo et al. / The Veterinary Journal 263 (2020) 105533 interaction was identified, Tukey’s post hoc procedure was used to compare least squares means between health status groups at each time point. Results Health status All cows in the RP group (n = 10) developed MET. The average detection date for MET was 6  2 DFC in RP and 8  3 in MET. Following the scoring system proposed by Urton et al. (2005), cows with RP had an average score of 3.2  0.8 and cows with MET had an average score of 2.5  0.8. No signs of lameness or other disorders were observed. Lying behaviour Patterns of daily lying time and lying time per bout are presented in Fig. 1A and B, respectively. Average daily lying time throughout the period under investigation was 691  26.4, 748  30.1, and 692  28.7 min/day (mean  SEM) for healthy cows, cows with RP, and cows with MET, respectively. Lying time remained at similar lengths from week –6 to week –1, was reduced between week –1 and week 1 (P < 0.01), then remained stable thereafter. Compared with healthy cows, the RP group had a longer lying time during most of the experimental period (HS*T; P < 0.01), especially in the 3 weeks before calving (P < 0.05). Lying time was not longer in the RP group than in healthy cows in the first week after calving (P = 0.06). Average lying time per bout was 74.8  3.8 min, with a marked reduction after calving (89.9  2.1 min before calving and 63.7  2.1 min after calving). Lying time was stable during the dry period until week –2, decreased from week –2 to week 1 (P < 0.01), then stable. This parameter did not differ overall among groups, but the interaction effect (HS*T; P < 0.01; Fig. 1B) revealed a Fig. 1. Least squares means ( standard error) of lying time (A; min/day) and lying per bout (B; min/bout) around calving in Holstein dairy cows grouped by health status: healthy (H, n = 13; cows without clinical diseases), cows diagnosed with retained placenta (RP, n = 10) and cows diagnosed with metritis (MET; n = 11). HS, overall effect of health status; T, overall effect of time (–14, –3, 1, 3, 7, and 28 DFC); HS*T, effect of interaction between health status and time. a–b Significant differences (P  0.05) between H, RP, and MET groups within each time point relative to calving. 3 slightly different pattern in healthy and sick cows. Cows with MET had longer lying bouts than other cows during the dry period (week -6 and -5; P < 0.05). Healthy cows had numerically longer lying bouts than sick cows after calving, but this difference was not statistically significant (P > 0.05). BCS, rectal temperature, and milk yield Patterns of BCS, rectal temperature, and milk yield are displayed in Fig. 2. Overall, statistically significant differences were not observed between groups for BCS, but the reduction in BCS was higher in cows that developed a disease, resulting in an interaction effect (HS*T; P = 0.02; Fig. 2A). The RP group had a more pronounced reduction in BCS than healthy cows or cows with MET. Health status did not affect overall rectal temperature. However, an interaction effect HS*T (P = 0.02; Table 2) revealed a different pattern for rectal temperature between disorder groups over time. Differences were statistically significant at 7 DFC, with a higher rectal temperature in cows with RP compared with healthy cows and those with MET (P < 0.01; Fig. 2B). No differences were detected between healthy cows and those with MET. Higher milk yield in the first 4 weeks of lactation was recorded for healthy cows compared with cows with RP and those with MET (HS*T; P = 0.05; Fig. 2C). Metabolic profiles Least squares means of plasma biomarkers are summarized in Table 2. Compared with healthy cows and those with MET, urea concentration was higher in cows with RP at 1 DFC (HS*T; P < 0.01; Fig. 3A). Compared with healthy cows, NEFA concentration was higher in cows with RP and those with MET at 1 DFC (HS; P = 0.05; Fig. 3B). At 3 DFC, cows with RP had higher BHB concentrations compared with healthy cows (P = 0.01), but not compared with those with MET (P = 0.07) Compared with healthy cows, magnesium concentration was lower in cows with RP and those with MET (HS; P < 0.05). The interaction HS*T effect (P < 0.05; Fig. 3C) demonstrated important differences between groups over time. The largest differences were observed after calving, with the lowest values in the disorder groups at 3 and 7 DFC compared with the H group (P < 0.01). No effect was detected for calcium, phosphorus, potassium, or chlorine (P > 0.05). AST-GOT concentrations were lower before calving in cows with RP and those with MET compared with healthy cows. Cows with RP had higher AST-GOT concentrations up to 7 DFC, then lower concentrations at 28 DFC, compared with cows with MET and healthy cows (HS*T; P < 0.01; Fig. 3D). Other liver function biomarkers were not affected (GGT, alkaline phosphatase, bilirubin, total protein; P > 0.05). Among positive acute-phase proteins, haptoglobin was higher in cows with RP and those with MET at 7 DFC (P < 0.01), and an interaction effect occurred (HS*T; P = 0.04; Fig. 3E). No differences were observed in ceruloplasmin concentrations (P > 0.05). Among the negative acute-phase proteins, compared with healthy cows, the increase in cholesterol concentrations in the first month of lactation was not statistically different in cows with RP and those with MET (HS*T; P = 0.06; Fig. 4C). Albumins and paraoxonase were lower for both groups at 7 DFC (HS*T; P < 0.01; Fig. 3F and 4B). Zinc concentrations were not different in the RP and MET groups compared with healthy cows (HS; P = 0.08; Fig. 4A). While there was no difference at 3 DFC when the RP and MET groups were compared with healthy cows (P = 0.10 and P = 0.09, respectively), there was a statistically significant reduction in plasma zinc concentration at –3 DFC for cows with MET (P = 0.01), compared with healthy cows. 4 L. Cattaneo et al. / The Veterinary Journal 263 (2020) 105533 Table 2 Least squares mean (LSM) of plasma biomarkers in periparturient Holstein cows from -14 to 28 days relative to calving with different health status: healthy (H, n = 13; cows without clinical diseases), cows diagnosed with retained placenta (RP, n = 10) and cows diagnosed with metritis (MET; n = 11). Health status (n = 34) P Biomarker H RP MET SEM HS T HS*T BCS Temperature,  C Hematocrit, L/L Glucose, mmol/La Urea, mmol/L NEFA, mmol/La BHB, mmol/La Creatinine, mmol/L Calcium, mmol/L Phosphorus, mmol/L Magnesium, mmol/L Sodium, mmol/L Potassium, mmol/L Chlorine, mmol/L Zinc, mmol/L Cholesterol, mmol/L Ceruloplasmin, mmol/L Albumin, g/L Globulin, g/L AST-GOT, U/La GGT, U/La Alkaline phosphatase, U/La Bilirubin, mmol/La Total protein, g/L Haptoglobin, g/L Paraoxonase, U/mL ROM, mgH2O2/100 m L SHp, mmol/L Myeloperoxidase, U/La FRAP, mmol/L TEa Retinol, mg/100 m L Tocoferol, mg/mLa β-carotene, mg/100 m L 2.82 38.5 0.34 4.11 4.69 0.49 0.62 92.6 2.44 1.52 1.03 148.1 4.20 109.0 12.4 2.92 2.82 36.1 40.6 101.3 24.6 57.4 4.03 76.7 0.49 73.8 15.0 309.6 479.9 159.2 26.8 2.44 0.18 2.78 38.7 0.32 4.24 5.01 0.63 0.74 96.1 2.42 1.52 0.95 150.1 4.30 110.0 11.9 2.71 2.75 35.5 39.2 100.4 21.5 47.9 4.42 74.7 0.54 70.8 14.6 304.2 482.9 149.2 24.2 1.49 0.14 2.81 38.6 0.32 4.14 4.92 0.54 0.64 92.9 2.39 1.64 0.95 148.6 4.19 108.9 11.0 2.71 2.84 35.3 43.1 99.7 22.3 59.4 4.66 78.5 0.61 65.9 15.0 304.6 473.5 142.2 24.0 1.96 0.18 0.07 0.24 0.01 0.24 0.45 0.15 0.17 3.4 0.08 0.17 0.04 1.2 0.19 1.1 1.1 0.34 0.25 0.9 2.8 9.4 2.9 10.4 1.48 2.3 0.12 6.6 1.0 16.9 29.0 13.6 3.7 0.61 0.04 <0.01 0.05 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.50 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.02 0.02 0.57 0.91 <0.01 0.72 0.28 0.08 0.95 0.43 <0.01 0.78 0.86 0.59 0.51 0.06 0.77 <0.01 0.31 <0.01 0.59 0.23 0.31 0.23 0.04 <0.01 0.71 0.10 0.25 0.36 0.02 0.12 0.70 0.78 0.28 0.01 0.35 0.47 0.05 0.57 0.34 0.66 0.35 <0.01 0.05 0.62 0.21 0.08 0.70 0.85 0.52 0.25 0.83 0.43 0.25 0.53 0.12 0.18 0.24 0.75 0.94 0.85 0.13 0.52 0.15 0.48 AST-GOT, aspartate aminotransferase–glutamate oxaloacetate transaminase; GGT, g-glutamyltransferase; NEFA, non esterified fatty acids; BHB, β-hydroxybutyrate; ROM, reactive oxygen metabolites; SHp, protein thiol groups; FRAP, ferric-reducing ability of plasma; SEM, highest standard error of the mean; HS, overall effect of health status; T, overall effect of time (-14, -3, 1, 3, 7, and 28 DFC); HS*T, effect of interaction between health status and time. a Log10 back-transformed LSM. Discussion Fig. 2. Least squares means ( standard error) of body condition score (BCS; A), rectal temperature (B;  C), and milk yield (C, Kg/day) around calving in Holstein dairy cows grouped by health status: healthy (H, n = 13; cows without clinical diseases), cows diagnosed with retained placenta (RP, n = 10), and cows diagnosed with metritis (MET; n = 11). HS, overall effect of health status; T, overall effect of time (–14, –3, 1, 3, 7, and 28 DFC); HS*T, effect of interaction between health status and time. a–b Significant differences (P  0.05) between H, RP, and MET groups within each time point relative to calving. Retinol was lower in cows with RP and those with MET at 7 DFC compared with healthy cows (HS*T; P = 0.02; Fig. 4D), whereas there was no difference in plasma tocopherol between groups. Oxidative stress biomarkers did not differ among groups (i.e., myeloperoxidase, thiol groups, FRAP, ROM, β-carotene). In this study, we investigated the relationship between lying behaviour before and after the diagnosis of a uterine disease during the transition period in Holstein cows. Automatic measurements of cow behaviour could be used as an early indicator to improve prevention and treatment of some diseases (Weary et al., 2009). In a recent study, Piñeiro et al. (2019) investigated the relationships between lying time, blood biomarkers, and diseases. In particular, lying time had a linear association with the risk of ketosis and a quadratic association with the concentration of NEFA at 7 and 14 days after calving in that study. In our study, in agreement with Kaufman et al. (2016), we observed that cows had similar patterns for daily lying time and lying time per bout around the time of calving, with higher values during the dry period (approximately 13 h/day), a sudden decrease approaching the time of calving, then a stabilization around the values recorded at calving (approximately 11 h/day). Hendriks et al. (2019) reported similar data for lying time around the time of calving for healthy grazing cows, but less time spent lying on average. These differences between dry and lactation periods could be attributed to the stress of calving and the huge changes that L. Cattaneo et al. / The Veterinary Journal 263 (2020) 105533 5 Fig. 3. Least squares means ( standard error) of significant plasma biomarkers around calving in Holstein dairy cows grouped by health status: healthy (H, n = 13; cows without clinical diseases), cows diagnosed with retained placenta (RP, n = 10), and cows diagnosed with metritis (MET; n = 11). a–c Significant differences (P  0.05) between H, RP, and MET groups within each time point relative to calving. occur in the transition period, such as lactogenesis, changes in management and environments between dry and lactating cows, and dietary modifications (Trevisi et al., 2016). Jensen et al. (2005) suggested that heifers required lying for 12 h/day. In our study, this threshold was exceeded by dry cows but was not reached by lactating cows. This could be explained by the time spent away from the pen for milking operations (up to 2 h/day) and the increased time spent eating, considering that lactating cows require twice the dry matter intake; both aspects reduce the available time for lying (Gomez and Cook, 2010; Huzzey et al., 2005). The average lying time during lactation in our study is comparable to the results of several published studies (Bewley et al., 2010; Ito et al., 2010; von Keyserlingk et al., 2012). To the best of our knowledge, this is the first study to investigate the relationship between lying time and RP. Until the first week after calving, cows that developed RP had a significantly longer lying time than others involved in this study. We hypothesized that this occurred because these cows were already experiencing illness, which decreased the time dedicated to feeding and increased time available for resting (Weary et al., 2009), even though rectal temperature was not elevated. Rodriguez-Jimenez et al. (2018) observed the same behaviour pattern (increased lying time and lower dry matter intake) in cows with subclinical ketosis. Further research on the link between increased LT and RP might reveal interesting information about the mechanisms involved. Our study was unable to demonstrate a statistically significant increase in the duration of lying bouts after calving when healthy cows were compared with sick cows. This could have been due to limitations in the statistical power of our study, perhaps due to insufficient group sizes for statistical comparisons. However, if the duration of each bout in healthy cows was longer, this could be suggestive of either a more relaxed demeanour or different feeding behaviour. Lomb et al. (2018) reported that healthy cows spent more of their standing time at the feed bunk eating, whereas cows with MET spent this time perching, standing in the stall, or in social or abnormal lying-related behaviours. Despite lower milk yields in sick cows in our study, this is also supported by the marked decrease in BCS in cows with RP, suggesting a lower feed intake compared with healthy cows. Feeding behaviour could be a relevant factor in our analysis because major changes occur during the peripartum period, especially in cows at increased risk of disease (von Keyserlingk and Weary, 2010). There was no difference in lying time between healthy cows and those with MET in our study, which is consistent with Barragan et al. (2018), who observed the same daily lying time in 6 L. Cattaneo et al. / The Veterinary Journal 263 (2020) 105533 Fig. 4. Least squares means ( standard error) of significant plasma biomarkers of negative acute-phase response around calving in Holstein dairy cows grouped by health status: healthy (H, n = 13; cows without clinical diseases), cows diagnosed with retained placenta (RP, n = 10), and cows diagnosed with metritis (MET; n = 11). a–b Significant differences (P  0.05) between H, RP, and MET groups within each time point relative to calving. multiparous cows with and without MET. In contrast, Neave et al. (2018) reported a reduction in lying time and fewer lying bouts before calving in cows diagnosed with MET, but no differences after calving. In our study, MET had a less pronounced impact on behaviour than RP. Based on work by Lomb et al. (2018), it is possible that the absence of an increase in lying time in our study, which usually occurs in response to sickness (Hart, 1988), was due to visceral pain associated with MET that could have caused an unwillingness to lie down. Cows with RP had a higher rectal temperature than healthy cows at 7 DFC. We could relate this increase to the inflammatory status caused by the disorder or the onset of subsequent uterine infection (Wenz et al., 2011). The absence of this increase in cows with MET may have been associated with the less severe infection status than in cows with RP, as demonstrated by lower uterine discharge score. Besides lying time, this data could also support the differences observed in milk yield. Lower production in sick cows is consistent with the findings of Rajala and Gröhn (1998) for cows with RP, and Huzzey et al. (2005) for cows with MET. Another reason for reduced milk production in the RP group could have been reduced feed intake. However, we cannot fully confirm this hypothesis because dry matter intake was not recorded in this study. We inferred such a response since these cows had a longer lying time up until the first week postpartum, which could have been related to less time spent eating. Sick animals typically devote less time to feeding, drinking, and reproduction, increasing the time for rest to conserve energy (Hart, 1988). This may also explain the larger reduction in BCS observed in cows with RP, which was not related to higher energy expenditure required for milk synthesis. The response in cows with RP may also have been modified by increased use of body fat, thereby reducing the difference in milk yield between groups. The analysis of the metabolic profile was carried out to assist and support behavioural observations and to evaluate potential differences in biomarker profiles between cows with or without a disorder in the periparturient period. As observed for BCS, the cows with RP and those with MET had increased mobilization of body fat immediately after calving. This result is supported by higher NEFA concentrations on the day after calving and by higher BHB concentrations at 3 DFC. We speculate that these cows spent less time eating and, consequently, had more reduced feed intake, thus relying more heavily on body reserves to support milk production. It is common for sick animals to exhibit anorexia, fever, lethargy, and reduced social activities (Dantzer and Kelley, 2007). Moreover, the mobilization of body fat can lead to immune dysfunction in periparturient dairy cows and could induce an inflammatory response (Sordillo and Raphael, 2013). Uterine health disorders also appear to be related to impaired function of peripheral blood neutrophils and negative energy balance (Hammon et al., 2006). Since cows with uterine disorders eat less, they have higher plasma NEFA and BHB, and consequently decreased neutrophil function before diagnosis. The lower magnesium concentration observed in sick cows in our study could also be explained by reduced ingestion since magnesium is usually well absorbed (Goff, 2008). Considering that diets were the same for all groups, this difference may have resulted from a lower feed intake. However, since the dry matter intake was not measured in this study, we cannot fully confirm our speculations. Uterine diseases such as RP and MET are known to cause a proinflammatory response. In this study, haptoglobin concentrations were higher in the unhealthy groups in the first week after calving. Plasma haptoglobin is an indicator of a positive acute-phase response that is associated with a reduction of negative acutephase proteins, which are usually synthesised in the liver (Bertoni et al., 2008). In this study, cows with RP and those with MET had lower concentrations of albumin, paraoxonase, and retinol. These observations support the presence of a mildly increased inflammatory status and impaired liver function (Bertoni et al., 2008). Although there was not a statistically significant effect of interaction between health status and time for plasma cholesterol, which was lower in sick cows over the study period, we speculate that recovery of liver function was not fully achieved in cows with L. Cattaneo et al. / The Veterinary Journal 263 (2020) 105533 RP at 28 DFC (Bertoni et al., 2008). The lower plasma zinc concentration in cows that developed a uterine disease could suggest a more severe inflammatory status. It has been reported that the plasma zinc concentration is reduced during the acutephase reaction because of increased liver synthesis of metallothionein, which sequesters zinc from blood to reduce pathogen survival (Rink and Kirchner, 2000; Trevisi and Minuti, 2018). In this study, retinol concentrations were lower in sick cows, especially at 7 DFC, suggesting a reduction of carrier proteins (retinol-binding proteins) as a result of the acute-phase response (Bertoni and Trevisi, 2013). In this context, the increased AST-GOT concentration observed in the cows with RP indicated that impairment of liver function occurred. Conclusions Cows later diagnosed with RP spent more time lying down during the prepartum period and in the first week after calving. The lying behaviour of the cows with MET was similar to those that remained healthy. These results suggest that automatic monitoring of lying time over the transition period may help identify multiparous cows at risk for RP, and has the potential to be used as a predictive tool in farm management. Associations between lying time and other parameters, such as rumination time and feeding behaviour, might help identify cows at increased risk of ill health. However, further studies with more statistical power are needed to determine reference ranges for lying time in healthy cows in different environmental conditions. Conflict of interest statement None of the authors has any financial or personal relationships that could inappropriately influence or bias the content of the paper. Acknowledgments This study was funded by the ‘Romeo ed Enrica Invernizzi foundation’, Milan, Italy. Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.tvjl.2020.105533. References Barragan, A.A., Piñeiro, J.M., Schuenemann, G.M., Rajala-Schultz, P.J., Sanders, D.E., Lakritz, J., Bas, S., 2018. Assessment of daily activity patterns and biomarkers of pain, inflammation, and stress in lactating dairy cows diagnosed with clinical metritis. Journal of Dairy Science 101, 8248–8258. Beagley, J.C., Whitman, K.J., Baptiste, K.E., Scherzer, J., 2010. Physiology and treatment of retained fetal membranes in cattle. Journal of Veterinary Internal Medicine 24, 261–268. Berckmans, D., 2014. Precision livestock farming technologies for welfare management in intensive livestock systems. Revue Scientifique Technique (International Office of Epizootics) 33, 189–196. Bertoni, G., Trevisi, E., 2013. Use of the Liver Activity Index and other metabolic variables in the assessment of metabolic health in dairy herds. Veterinary Clinics of North America Food Animal Practice 29, 413–431. Bertoni, G., Trevisi, E., Han, X., Bionaz, M., 2008. Effects of inflammatory conditions on liver activity in Puerperium Period and consequences for performance in dairy cows. Journal of Dairy Science 91, 3300–3310. Bewley, J.M., Boyce, R.E., Hockin, J., Munksgaard, L., Eicher, S.D., Einstein, M.E., Schutz, M.M., 2010. Influence of milk yield, stage of lactation, and body condition on dairy cattle lying behaviour measured using an automated activity monitoring sensor. Journal of Dairy Research 77, 1. Calamari, L., Soriani, N., Panella, G., Petrera, F., Minuti, A., Trevisi, E., 2014. Rumination time around calving: an early signal to detect cows at greater risk of disease. Journal of Dairy Science 97, 3635–3647. 7 Cook, N.B., 2019. Optimizing resting behavior in lactating dairy cows through freestall design. Veterinary Clinics of North America Food Animal Practice, Housing to Optimize Comfort. Health and Productivity of Dairy Cattle 35, 93– 109. Cook, N.B., Mentink, R.L., Bennett, T.B., Burgi, K., 2007. The effect of heat stress and lameness on time budgets of lactating dairy cows. Journal of Dairy Science 90, 1674–1682. Cooper, M.D., Arney, D.R., Phillips, C.J.C., 2007. Two- or four-hour lying deprivation on the behavior of lactating dairy cows. Journal of Dairy Science 90, 1149–1158. Dantzer, R., Kelley, K.W., 2007. Twenty years of research on cytokine-induced sickness behavior. Brain Behavior and Immunity 21, 153–160. DeVries, T.J., von Keyserlingk, M.A.G., 2005. Time of feed delivery affects the feeding and lying patterns of dairy cows. Journal of Dairy Science 88, 625–631. Fregonesi, J.A., Tucker, C.B., Weary, D.M., 2007. Overstocking reduces lying time in dairy cows. Journal of Dairy Science 90, 3349–3354. Giuliodori, M.J., Magnasco, R.P., Becu-Villalobos, D., Lacau-Mengido, I.M., Risco, C.A., de la Sota, R.L., 2013. Metritis in dairy cows: risk factors and reproductive performance. Journal of Dairy Science 96, 3621–3631. Goff, J.P., 2008. The monitoring, prevention, and treatment of milk fever and subclinical hypocalcemia in dairy cows. The Veterinary Journal 176, 50–57. Gomez, A., Cook, N.B., 2010. Time budgets of lactating dairy cattle in commercial freestall herds. Journal of Dairy Science 93, 5772–5781. Grant, R., 2007. Taking advantage of natural behavior improves dairy cow performance. Proceedings of Western Dairy Management Conference, Reno, NV, pp. 225–236. Halachmi, I., Guarino, M., Bewley, J., Pastell, M., 2019. Smart animal agriculture: application of real-time sensors to improve animal well-being and production. Annual Review of Animal Bioscience 7, 403–425. Hammon, D.S., Evjen, I.M., Dhiman, T.R., Goff, J.P., Walters, J.L., 2006. Neutrophil function and energy status in Holstein cows with uterine health disorders. Veterinary Immunology and Immunopathology 113, 21–29. Hart, B.L., 1988. Biological basis of the behavior of sick animals. Neuroscience & Biobehavioral Reviews 12, 123–137. Hendriks, S.J., Phyn, C.V.C., Turner, S.-A., Mueller, K.M., Kuhn-Sherlock, B., Donaghy, D.J., Huzzey, J.M., Roche, J.R., 2019. Lying behavior and activity during the transition period of clinically healthy grazing dairy cows. Journal of Dairy Science 102, 7371–7384. Henriksen, J.C., Munksgaard, L., 2019. Validation of AfiTagII, a device for automatic measuring of lying behaviour in Holstein and Jersey cows on two different bedding materials. Animal 13, 617–621. Hillman, P.E., Lee, C.N., Willard, S.T., 2005. Thermoregulatory responses associated with lying and standing in heat-stressed dairy cows. Transactions of the American Society of Agricultural Engineers 48, 795–801. Huzzey, J.M., von Keyserlingk, M.A.G., Weary, D.M., 2005. Changes in feeding, drinking, and standing behavior of dairy cows during the transition period. Journal of Dairy Science 88, 2454–2461. Ito, K., von Keyserlingk, M.A.G., LeBlanc, S.J., Weary, D.M., 2010. Lying behavior as an indicator of lameness in dairy cows. Journal of Dairy Science 93, 3553–3560. Jensen, M.B., Proudfoot, K.L., 2017. Effect of group size and health status on behavior and feed intake of multiparous dairy cows in early lactation. Journal of Dairy Science 100, 9759–9768. Jensen, M.B., Pedersen, L.J., Munksgaard, L., 2005. The effect of reward duration on demand functions for rest in dairy heifers and lying requirements as measured by demand functions. Applied Animal Behaviour Science 90, 207–217. Kaufman, E.I., LeBlanc, S.J., McBride, B.W., Duffield, T.F., DeVries, T.J., 2016. Short communication: association of lying behavior and subclinical ketosis in transition dairy cows. Journal of Dairy Science 99, 7473–7480. Laven, R.A., Peters, A.R., 1996. Bovine retained placenta: aetiology, pathogenesis and economic loss. Veterinary Record 139, 465–471. Littell, R.C., Henry, P.R., Ammerman, C.B., 1998. Statistical analysis of repeated measures data using SAS procedures. Journal of Animal Science 76, 1216–1231. Lomb, J., Weary, D.M., Mills, K.E., von Keyserlingk, M.A.G., 2018. Effects of metritis on stall use and social behavior at the lying stall. Journal of Dairy Science 101, 7471– 7479. Lopreiato, V., Minuti, A., Cappelli, F.P., Vailati-Riboni, M., Britti, D., Trevisi, E., Morittu, V.M., 2018. Daily rumination pattern recorded by an automatic rumination-monitoring system in pre-weaned calves fed whole bulk milk and ad libitum calf starter. Livestock Science 212, 127–130. Lopreiato, V., Minuti, A., Trimboli, F., Britti, D., Morittu, V.M., Cappelli, F.P., Loor, J.J., Trevisi, E., 2019. Immunometabolic status and productive performance differences between periparturient Simmental and Holstein dairy cows in response to pegbovigrastim. Journal of Dairy Science 102, 9312–9327. Lopreiato, V., Vailati-Riboni, M., Morittu, V.M., Britti, D., Piccioli-Cappelli, F., Trevisi, E., Minuti, A., 2020. Post-weaning rumen fermentation of Simmental calves in response to weaning age and relationship with rumination time measured by the Hr-Tag rumination-monitoring system. Livestock Science 232, 103918. Maselyne, J., Pastell, M., Thomsen, P.T., Thorup, V.M., Hänninen, L., Vangeyte, J., Van Nuffel, A., Munksgaard, L., 2017. Daily lying time, motion index and step frequency in dairy cows change throughout lactation. Research in Veterinary Science 110, 1–3. Mezzetti, M., Minuti, A., Piccioli-Cappelli, F., Amadori, M., Bionaz, M., Trevisi, E., 2019. The role of altered immune function during the dry period in promoting the development of subclinical ketosis in early lactation. Journal of Dairy Science 102, 9241–9258. Moretti, P., Probo, M., Morandi, N., Trevisi, E., Ferrari, A., Minuti, A., Venturini, M., Paltrinieri, S., Giordano, A., 2015. Early post-partum hematological changes in 8 L. Cattaneo et al. / The Veterinary Journal 263 (2020) 105533 Holstein dairy cows with retained placenta. Animal Reproduction Science 152, 17–25. Munksgaard, L., Jensen, M.B., Pedersen, L.J., Hansen, S.W., Matthews, L., 2005. Quantifying behavioural priorities—effects of time constraints on behaviour of dairy cows, Bos taurus. Applied Animal Behaviour Science 92, 3–14. National Research Council, 2001. Nutrient Requirements of Dairy Cattle, 7th ed. Washington, DC. Neave, H.W., Lomb, J., von Keyserlingk, M.A.G., Behnam-Shabahang, A., Weary, D.M., 2017. Parity differences in the behavior of transition dairy cows. Journal of Dairy Science 100, 548–561. Neave, H.W., Lomb, J., Weary, D.M., LeBlanc, S.J., Huzzey, J.M., von Keyserlingk, M.A. G., 2018. Behavioral changes before metritis diagnosis in dairy cows. Journal of Dairy Science 101, 4388–4399. Peeler, E.J., Otte, M.J., Esslemont, R.J., 1994. Inter-relationships of periparturient diseases in dairy cows. The Veterinary Record 134, 129–132. Piñeiro, J.M., Menichetti, B.T., Barragan, A.A., Relling, A.E., Weiss, W.P., Bas, S., Schuenemann, G.M., 2019. Associations of pre- and postpartum lying time with metabolic, inflammation, and health status of lactating dairy cows. Journal of Dairy Science 102, 3348–3361. Rajala, P.J., Gröhn, Y.T., 1998. Effects of Dystocia, retained placenta, and Metritis on milk yield in dairy cows. Journal of Dairy Science 81, 3172–3181. Rink, L., Kirchner, H., 2000. Zinc-altered immune function and cytokine production. The Journal of Nutrition 130, 1407S–1411S. Rodriguez-Jimenez, S., Haerr, K.J., Trevisi, E., Loor, J.J., Cardoso, F.C., Osorio, J.S., 2018. Prepartal standing behavior as a parameter for early detection of postpartal subclinical ketosis associated with inflammation and liver function biomarkers in peripartal dairy cows. Journal of Dairy Science 101, 8224–8235. Sepúlveda-Varas, P., Weary, D.M., von Keyserlingk, M.A.G., 2014. Lying behavior and postpartum health status in grazing dairy cows. Journal of Dairy Science 97, 6334–6343. Siivonen, J., Taponen, S., Hovinen, M., Pastell, M., Lensink, B.J., Pyörälä, S., Hänninen, L., 2011. Impact of acute clinical mastitis on cow behaviour. Applied Animal Behaviour Science 132, 101–106. Solano, L., Barkema, H.W., Pajor, E.A., Mason, S., LeBlanc, S.J., Nash, C.G.R., Haley, D.B., Pellerin, D., Rushen, J., de Passillé, A.M., Vasseur, E., Orsel, K., 2016. Associations between lying behavior and lameness in Canadian Holstein-Friesian cows housed in freestall barns. Journal of Dairy Science 99, 2086–2101. Sordillo, L.M., Raphael, W., 2013. Significance of metabolic stress, lipid mobilization, and inflammation on transition cow disorders. Veterinary Clinics of North America Food Animal Practice 29, 267–278. Steensels, M., Bahr, C., Berckmans, D., Halachmi, I., Antler, A., Maltz, E., 2012. Lying patterns of high producing healthy dairy cows after calving in commercial herds as affected by age, environmental conditions and production. Applied Animal Behaviour Science 136, 88–95. Trevisi, E., Minuti, A., 2018. Assessment of the innate immune response in the periparturient cow. Research in Veterinary Science 116, 47–54. Trevisi, E., Moscati, L., Amadori, M., 2016. Disease-predicting and prognostic potential of innate immune responses to noninfectious stressors: human and animal models. The Innate Immune Response to Noninfectious Stressors. Academic Press, Cambridge, MA, pp. 209–235. Urton, G., von Keyserlingk, M.A.G., Weary, D.M., 2005. Feeding behavior identifies dairy cows at risk for metritis. Journal of Dairy Science 88, 2843–2849. von Keyserlingk, M.A.G., Weary, D.M., 2010. Review: feeding behaviour of dairy cattle: meaures and applications. Canadian Journal of Animal Science 90, 303– 309. von Keyserlingk, M.A.G., Barrientos, A., Ito, K., Galo, E., Weary, D.M., 2012. Benchmarking cow comfort on North American freestall dairies: lameness, leg injuries, lying time, facility design, and management for high-producing Holstein dairy cows. Journal of Dairy Science 95, 7399–7408. Weary, D.M., Huzzey, J.M., von Keyserlingk, M.A.G., 2009. Board-Invited Review: using behavior to predict and identify ill health in animals. Journal of Animal Science 87, 770–777. Wenz, J.R., Moore, D.A., Kasimanickam, R., 2011. Factors associated with the rectal temperature of Holstein dairy cows during the first 10 days in milk. Journal of Dairy Science 94, 1864–1872. Westin, R., Vaughan, A., de Passillé, A.M., DeVries, T.J., Pajor, E.A., Pellerin, D., Siegford, J.M., Vasseur, E., Rushen, J., 2016. Lying times of lactating cows on dairy farms with automatic milking systems and the relation to lameness, leg lesions, and body condition score. Journal of Dairy Science 99, 551–561.