Hypocalcemia in dairy cows: meta-analysis and dietary cation anion difference theory revisited.

Data from 137 published trials involving 2,545 calvings were analyzed using random effects normal logistic regression models to identify risk factors for clinical hypocalcemia in dairy cows. The aim of the study was to examine which form, if any, of the dietary cation anion difference (DCAD) equation provided the best estimate of milk fever risk and to clarify roles of calcium, magnesium, and phosphorus concentrations of prepartum diets in the pathogenesis of milk fever. Two statistically equivalent and biologically plausible models were developed that predict incidence of milk fever. These models were validated using data from 37 trials excluded from the original data used to generate the models; missing variables were replaced with mean values from the analyzed data. The preferred models differed slightly; Model 1 included prepartum DCAD, and Model 2 included prepartum dietary concentrations of potassium and sulfur alone, but not sodium and chloride. Other factors, included in both models were prepartum dietary concentrations of calcium, magnesium, phosphorus; days exposed to the prepartum diet; and breed. Jersey cows were at 2.25 times higher risk of milk fever than Holstein cows in Model 1. The results support the DCAD theory of greater risk of milk fever with higher prepartum dietary DCAD (odds ratio = 1.015). The only DCAD equation supported in statistical analyses was (Na(+) + K(+)) - (Cl(-) + S(2-)). This finding highlights the difference between developing equations to predict DCAD and those to predict milk fever. The results support a hypothesis of a quadratic role for Ca in the pathogenesis of milk fever (model 1, odds ratio = 0.131; Model 2, odds ratio = 0.115). Milk fever risk was highest with a prepartum dietary concentration of 1.35% calcium. Increasing prepartum dietary magnesium concentrations had the largest effect on decreasing incidence of milk fever in both Model 1 (odds ratio = 0.006) and Model 2 (odds ratio = 0.001). Increasing dietary phosphorus concentrations prepartum increased the risk of milk fever (Model 1, odds ratio = 6.376; Model 2, odds ratio = 9.872). The models presented provide the basis for the formulation of diets to reduce the risk of milk fever and strongly support the need to evaluate macro mineral nutrition apart from DCAD of the diet.