18/05/2022
Periparturient hypocalcemia is a common metabolic disorder in dairy cows that leads to an increased risk of detrimental health and production outcomes and in severe cases can be life threatening. Physiologically, serum calcium concentration in the adult cow is maintained above 2.0 mmol/L (Martin-Tereso and Martens, 2014). Due to the start of colostrum production and consequently increasing calcium demand, the nadir of serum calcium concentration occurs 12 to 24 h after parturition (Kimura et al., 2006; Goff, 2008).
Hypocalcemia is considered as a gateway disease and predisposes the cow to various metabolic and infectious disorders in early lactation (Goff, 2008) such as metritis (Martinez et al., 2012) and mastitis (Curtis et al., 1983). In a study by Martinez et al. (2012), numbers of neutrophils were reduced and their ability to undergo phagocytosis and oxidative burst was impaired in cows affected by hypocalcemia which might in part explain the increased risk for infectious diseases. On a cellular level, suppressed function of immune cells was mediated by reduced cytosolic calcium concentration (Martinez et al., 2014).
Cows with naturally occurring hypocalcemia at parturition had elevated concentrations of NEFA and BHBA as indicators of increased lipomobilization (Martinez et al., 2012). The same group of authors were able to repeat these finding in cows with induced hypocalcemia (Martinez et al., 2014). Induction of hypocalcemia with EDTA infusion caused reduced DMI and decreased plasma concentrations of insulin. These negative effects are supported by other studies showing an increased risk for displaced abomasum (Chapinal et al., 2011; Seifi et al., 2011), increased weight loss in early lactation (Caixeta et al., 2015), and ultimately an increased culling risk (Seifi et al., 2011; Roberts et al., 2012) for cows with hypocalcemia. Furthermore, subclinical hypocalcemia affected reproductive performance such as estrous cyclicity (Ribeiro et al., 2013; Caixeta et al., 2017) and pregnancy rate to first AI (Chapinal et al., 2012).
In a retrospective study including 1,462 cows from 480 dairy farms in 21 states of the United States, the prevalence of hypocalcemia was 25% in first lactation cows and about 50% in multiparous cows (Reinhardt et al., 2011). Clinical milk fever was prevalent in 1, 4, 6, and 10% of first, second, third, and ≥fourth lactation cows, respectively. These results originate from the 2002 National Animal Health Monitoring System (NAHMS) dairy study (USDA, 2002). This study has been used as a reference for the prevalence of hypocalcemia multiple times. But the study was not specifically designed to estimate cow- and herd-level prevalence of hypocalcemia and the currentness of the results are limited.
More recently, different strategies (e.g., oral calcium supplementation, anionic salts) to prevent hypocalcemia have evolved and were implemented in the dairy industry (Martin-Tereso and Martens, 2014). These approaches might affect the prevalence of hypocalcemia. To our knowledge, however, no information is available reporting the actual prevalence of hypocalcemia and associated preventive strategies.
Therefore, the objective of this study was to estimate the prevalence of hypocalcemia on a cow level and the implemented preventive strategies to control for hypocalcemia in commercial German dairy herds.
MATERIALS AND METHODS
The experimental procedures reported herein were conducted with the approval of the Institutional Animal Care and Use Committee of the Freie Universität Berlin.
Study Population
A cross-sectional study was conducted based on a convenience sample of 115 dairy herds from 8 federal states of Germany between February 2015 and August 2016. Inclusion criteria for herds were (1) participation in a federal DHIA testing system, (2) freestall housing with at least 100 milking cows, (3) feeding of a TMR-based diet, and (4) a computerized herd management software. Average herd size was 513 and ranged from 112 to 2,607 lactating cows. The average milk production (305-d ECM, 4.0% fat, 3.4% protein) was 9,231 kg (range 6,257–10,880 kg). Holstein Friesian cows were the dominant breed on 112 farms. Two farms kept Simmental cattle and 1 farm Jersey as the dominant breed.
A sample size calculation was conducted according to Dohoo et al. (2009). We assumed that the prevalence of milk fever tends to be highly clustered within herds because of the effect of herd management (e.g., breed, dry cow nutrition) on the risk of hypocalcemia. Therefore, we selected an intra-cluster correlation coefficient of 0.3. A sample size of 1,388 animals with 12 animals per herd was deemed adequate to estimate the true prevalence of subclinical hypocalcemia on a cow level with 95% confidence and 10% precision.
If a farm provided less than 12 blood samples the farm was excluded from statistical analysis. If a farm supplied more than 12 blood samples, 12 cows were selected, using a random function in Excel (Office 2010, Microsoft Deutschland Ltd., Munich, Germany). A random list was generated separately for each of the farms.
Experimental Procedures
Veterinary practitioners had been invited to participate in the study by an information leaflet sent out by regular mail. Participating practices were informed about the nature and duration of the study and received a package containing serum blood collection systems (S-Monovette 9mL Z, Sarstedt AG and Co, Nürnbrecht, Germany), cryo-vials (Cryvial, Carl Roth GmbH and Co. KG, Karlsruhe, Germany) to store serum at −20°C until analysis, and a written standard operating procedure. This standard operating procedure described which information to record for each cow enrolled and how to examine the cow before blood collection. A case report form for each cow was provided to document time of sampling, ear tag number, time of calving, calving ease (i.e., unassisted calving or assisted calving with at least one person), clinical symptoms of milk fever (i.e., recumbency), and parity. Administration of calcium products, time relative to calving, and route of administration (i.e., subcutaneous, intravenous, oral) of these products was also documented. Sampling 12 cows per herd, veterinarians were asked to include 4 primiparous cows into the cohort. The farm personnel was asked, if other preventive strategies, such as feeding of anionic salts in the close-up group or injection of vitamin D before calving were implemented.
Animals were enrolled by convenience when a veterinarian visited the farm on a given day and an animal met the inclusion criteria of being within 48 h after parturition.
Definition of Hypocalcemia on Cow Level and Herd Level
Normocalcemia was defined as serum calcium concentration greater or equal to 2.0 mmol/L (Reinhardt et al., 2011). Cows not affected clinically but with a serum calcium concentration below 2.0 mmol/L were categorized as subclinical hypocalcemic animals. Recumbent cows with a serum calcium concentration below 2.0 mmol/L were defined as cows suffering from clinical milk fever. Although often used in current literature (Reinhardt et al., 2011; Wilhelm et al., 2017) it was recently shown that 2.0 mmol/L is the most conservative approach as higher thresholds (i.e., 2.1 and 2.2 mmol/L) were also associated with negative health or production outcomes (Chapinal et al., 2011; Seifi et al., 2011; Chapinal et al., 2012; Martinez et al., 2012; Roberts et al., 2012). Therefore, analyses were conducted considering 3 thresholds (i.e., 2.0, 2.1, and 2.2 mmol/L).
Based on the results of the sampled cohort per farm, herds were categorized as negative (≤2 animals per herd with serum calcium concentration below the threshold), borderline (3 to 5 animals per herd with serum calcium concentration below the threshold), or positive (≥6 animals per herd with serum calcium concentration below the threshold) according to Cook et al. (2006). Using a confidence level of 75% and an alarm level of 30%, the sampling of 12 animals per herd is adequate to classify herds into 3 categories Classification of blood calcium concentrations using 75% CI and an alarm level of 30% for test results from 12 cows sampled from a group of 100 cows. This calculation illustrates the association between positive blood samples in the cohort and prevalence of hypocalcemia in the tested herd.
Blood Sampling and Laboratory Analyses
Blood samples were taken from the coccygeal vessels using a serum blood collection system. Samples were kept at room temperature and allowed to clot. Within 5 h of blood collection, samples were centrifuged to harvest serum, which was frozen at −20°C. Analysis of blood samples was carried out by a commercial laboratory (Synlab Services GmbH, Augsburg, Germany). Total serum calcium, magnesium, and phosphorus concentration was analyzed using photometry (AU680, Beckman Coulter, Krefeld, Germany). The interassay coefficient of variation was 1.03% (Ca 2.37 mmol/L; n = 16), 1.06% (Mg = 0.99 mmol/L; n = 16), and 2.43% (P = 0.83 mmol/L; n = 16) for calcium, magnesium, and phosphorus, respectively. The intraassay coefficient of variation was 1.19% (Ca 2.40 mmol/L; n = 10), 0.88% (Mg = 0.99 mmol/L; n = 10), and 1.03% (P = 0.85 mmol/L; n = 10) for calcium, magnesium, and phosphorus, respectively.
Statistical Analyses
Individual cow data were transferred to Microsoft Excel (Office 2013, Microsoft Deutschland Ltd.). Statistical analyses were performed using SPSS for Windows (version 22.0, SPSS Inc., IBM, Ehningen, Germany). The association of lactation number and the type of hypocalcemia or the type of individual preventive strategy was analyzed using cross tabulations and χ2 tests.
For evaluation of the association between serum calcium concentration and time of calving, ease of calving, and time interval from calving to sampling, we used the GENLINMIXED procedure of SPSS. Cow was the experimental unit and herd was considered as a random effect. According to the model-building strategies described by Dohoo et al. (2009), each parameter considered for the mixed model should be separately analyzed in a univariate model, including the parameter as a fixed factor (i.e., categorical parameter) or covariate (i.e., continuous parameter). Only parameters resulting in univariate models with P ≤ 0.2 should be included in the final mixed model. The initial model contained the following explanatory variables as fixed effects: parity (1, 2, 3, ≥4), breed (Holstein, Jersey, or Simmental), time of calving (daytime from 0600 to 1759 h vs. nighttime 1800 to 0559 h), calving ease (unassisted calving vs. assisted calving), and time interval from calving to sampling (continuous; 0 to 48 h).
The concentration of calcium was related to the concentration of phosphorus or magnesium using a linear regression model and the LINEAR REGRESSION procedure from SPSS: yi = a + bXi, where yi is the dependent variable (magnesium or phosphorous concentration), Xi is the independent variable (calcium concentration), b is the slope of the regression line, and a is the intercept. R2 describes the coefficient of determination, which is the relative proportion of variance in yi that can be explained
Overall, blood samples were drawn from 1,709 animals at 0 to 48 h after calving from 125 farms. Ten farms with 54 animals were excluded from analysis because they provided less than 12 samples per farm. Another 275 animals were randomly excluded because 60 farms provided more than 12 samples per farm.
Data of 1,380 animals were available for final analyses. Of those, 228 (16.5%), 355 (25.7%), 332 (24.1%), and 465 (33.7%) were in first, second, third, and ≥fourth lactation, respectively.
Based on a calcium threshold of 2.0 mmol/L and clinical signs, the prevalence of subclinical hypocalcemia and clinical milk fever was 40.7% (561/1,380) and 7.2% (99/1,380), respectively. Considering higher thresholds of 2.1 and 2.2 mmol/L, prevalence of subclinical hypocalcemia increased to 53.0% (732/1380) and 67.5% (931/1380), respectively The prevalence of hypocalcemia increased with parity. None of the cows in first lactation was suffering from clinical milk fever. Prevalence of clinical milk fever was 1.4% (5/355), 5.7% (19/332), and 16.1% (75/465) for second, third, and ≥fourth parity cows, respectively
Table 1. Prevalence (no./total; % in parentheses) of subclinical hypocalcemia and milk fever 0 to 48 h after parturition in dairy cows stratified by parity considering 3 thresholds for blood calcium
Type of hypocalcemia Lactation 1 Lactation 2 Lactation 3 Lactation ≥4
Subclinical
Threshold 2.0 mmol/L 13/228a 103/355b 164/332c 281/465d
(5.7) (29.0) (49.4) (60.4)
Threshold 2.1 mmol/L 32/228a 158/355b 211/332c 331/465d
(14.0) (44.5) (63.6) (71.2)
Threshold 2.2 mmol/L 83/228a 222/355b 256/332c 370/465bc
(36.4) (62.5) (77.1) (79.6)
Clinical 0/228a 5/355a 19/332b 75/465c
(0.0) (1.4) (5.7) (16.1)
a–d
Values with different superscripts within rows differ, P < 0.05.
A significant effect of parity (P < 0.001) was observed on serum calcium concentration. Cows in first, second, third, or ≥fourth lactation had a serum calcium concentration of 2.213 mmol/L (95% CI: 2.054–2.372), 2.102 mmol/L (95% CI: 1.942–2.262), 1.997 mmol/L (95% CI: 1.837–2.158), and 1.891 mmol/L (95% CI: 1.735–2.046), respectively. Serum calcium concentration of cows that calved at night was 0.087 mmol/L higher compared with cows that calved during the day (95% CI: 0.044–0.130; P = 0.001). No significant effect of breed (P = 0.811), time from calving to sampling (P = 0.288), or calving ease (P = 0.902) was observed on serum calcium concentration.
A negative association of serum calcium and serum magnesium concentration was observed (y = −0.208x + 1.454; R2 = 0.151; P < 0.001; In contrast, there was a positive association of serum calcium and serum phosphorus concentration (y = 0.900x − 0.229; R2 = 0.335; P < 0.001; Association between serum calcium and serum magnesium concentration for all cows (n = 1,380, y = −0.208x + 1.454; R2 = 0.151; P < 0.001).
Association between serum calcium and serum phosphorus concentration for all cows (n = 1,380, y = 0.900x – 0.229; R2 = 0.335; P < 0.001).
Prevention of hypocalcemia on a cow level was more prevalent in multiparous cows compared with primiparous cows. Oral calcium supplementation was implemented in 13.8, 24.1, and 26.0% in second, third, and ≥fourth parity cows (Table 2) and most prevalent for prevention of hypocalcemia at the cow-level (n = 255; Table 2). Subcutaneous calcium injection or prepartum vitamin D application played a minor role and was on the herd level always combined with another preventive strategy (Table 2). In 34.8% (40/115), 6.1% (7/115), and 2.6% (3/115) of the herds, oral calcium supplementation, anionic salts, or a combination of both was used to control hypocalcemia on a herd level, respectively. Most of the herds (65/115) did not implement a control strategy for hypocalcemia.
Type of prophylaxis (no./total; % in parentheses) to prevent hypocalcemia in 1,380 cows of 115 dairy herds considering parity
Type of prophylaxis Lactation 1 Lactation 2 Lactation 3 Lactation ≥4
Oral calcium 5/228a 49/355b 80/332c 121/465c
(2.2) (13.8) (24.1) (26.0)
Subcutaneous calcium 0/228a 7/355ab 14/332b 23/465b
(0.0) (2.0) (4.2) (4.9)
Vitamin D prepartum 0/228ab 0/355b 7/332ac 15/465c
(0.0) (0.0) (2.1) (3.2)
Overall 5/228 56/355 101/332 159/465
(2.2) (15.8) (30.4) (34.2)
a–c
Values with different superscripts within rows differ, P < 0.05.
Only 12.2% of the 115 herds enrolled were classified as negative based on the alarm levels set by Cook et al. (2006).