Ketosis in Dairy Cows during Early Lactation - Detection in Pooled Blood Serum Samples

Abstract

Background: Ketosis is the most important metabolic disease with prevalence from 15 to 45%. Ketosis is diagnosed using a metabolic profile. Due to the high prevalence, it is necessary to determine a large number of metabolic profiles within farm, which represents an additional cost, so the implementation of pooled serum in assessing the metabolic status of cows was examined. The aim of this study was to validate and evaluate the influence of the relative position (Z-score) of the value of pooled sample metabolic parameters within the known reference value of healthy cows in the detection of ketosis in herd during early lactation.

Materials, Methods & Results: The experiment has been carried out using  blood samples collected by puncture of coccygeal vein from 50 ketotic and 50 healthy cows. Laboratory analysis includes determination of beta-hydroxybutyrate-BHB, non-esterified fatty acids-NEFA, cholesterol-CHOL, triglycerides-TGC, glucose-GLU, albumin-ALB, total protein-TPROT, UREA, Ca, P, total bilirubin-TBIL and aspartat aminotransferase-AST. The pooled serum was made from 10 individual samples originating from 10 different cows. A serum aliquot of 0.1 mL was taken from each sample, and a 1 mL volume of pooled serum was finally formed. Three types of serum pools were made: 1) 30 pooled sample were from ketosis; 2) 30 pooled sample were from healthy cows and 3) 60 pooled samples containing mixed sera of healthy cows and cows with ketosis were made as follows: 10 pools contain 10% to 60% of ketotic cows (1/10 to 6/10 samples). Statistical analysis includes: a) difference in metabolite concentration and Z-score in pooled sample and arithmetic mean individual sample in healthy and ketotic cow, b) correlation between Z-score of pooled sample and arithmetic mean of individual sample, c) ability of Z-score of metabolite to divide ketotic from healthy cow, d) correlation between Z-score and % of ketotic cow in pooled sample; and e) calculation of 95%CI of pooled sample Z-scores for each % of ketotic cow in pools. Z-score and all analysis were calculated for each metabolic parameter. The results of the study show that the mean values and Z-scores of the pool and the calculated average value of the individual samples participating in that pool differ significantly in healthy cows and cows in ketosis, except for TPROT and Ca. A higher value and a higher Z-score were found for BHB, NEFA, UREA, TBIL and AST, and a lower value and a lower Z-score for TGC, CHOL, GLU, ALB and P in ketotic cows compared to healthy cows. The value of the Z-score of the pooled sample and the calculated mean values of individual samples participating in the pool are highly correlated with each other (coefficient of determination over 99%). Z-score of metabolites in the pooled sample can be used to distinguish healthy from ketotic cows (ROC AUC= 0.711 to 0.989), except for TPROT and Ca. The Z-score value of the pooled sample shows a linear correlation with the percentage of ketotic cows in the pool and the reference ranges of Z-scores change significantly as a function of the percentage of ketosis cows.

Discussion: Modern research on the metabolic profile in cows requires obtaining a large amount of information from as few samples as possible. The advantages of using the Z-score are reflected in the following: this score does not depend on the absolute value of the metabolite, but on the position within the known population reference value, Z-score of sample and the arithmetic mean of individual samples included in the same pool are almost identical, the Z-score of these 2 groups of results is ideally correlated, the Z-score significantly correlates with the % of ketosis samples in the pooled sample. The use of pooled sample Z-score can be a useful in a herd level assessment of metabolic status and detection of ketosis as most important metabolic disease in dairy cows.

Keywords: dairy cattle, ketosis, metabolic disease, metabolic profile, pooled serum, z-score, diagnostics.