Validation of Brix refractometers and a hydrometer for measuring the quality of caprine colostrum

On-farm assessment of caprine colostrum quality is important for goat farmers; the ability to quickly recognize whether colostrum is suitable to feed to kids helps achieve successful passive transfer of immunity. The study compared the use of optical and digital Brix refractometers and a hydrometer against the international gold standard radial immunodiffusion (RID), using both fresh and frozen samples. A locally available ELISA methodology was included for comparison. A total of 300 samples were collected from 2 farms (farm 1: n = 157, collected by research staff within 24 h of parturition; farm 2: n = 143, collected by the farmer within 12 h of parturition). Farm 1 provided doe age for a subset of samples (n = 86). Samples were tested fresh and then frozen for shipment and repeated testing. Specific gravity was measured using a hydrometer in a subset of samples (n = 22) from farm 2. Because no gold standard thresholds are currently available for caprine colostrum, RID-derived values of 30, 40, and 50 g/L IgG were used as potential "good quality" thresholds. Pearson (ρ) and Lin's concordance correlation coefficients (CCC) were calculated for comparison of methods. Optimum thresholds were established maximizing the Youden index and minimizing the "distance closest to the top left corner" of the receiver operator characteristic curves. Brix values were correlated with RID (optical Brix, fresh: ρ = 0.73; digital Brix, fresh: ρ = 0.71; digital Brix, frozen: ρ = 0.76) and with each other (range: ρ = 0.93 to 0.99; CCC = 0.91 to 0.99). Specific gravity measured by the hydrometer yielded a strong relationship with RID (ρ = 0.83) and with Brix values (range: ρ = 0.88 to 0.90). The ELISA method was not correlated with Brix methods (range: ρ = 0.02 to 0.09) or RID (ρ = 0.20). Depending on the colostrum IgG threshold, the hydrometer yielded high Youden indices (range: 0.78 to 0.93) and low distance closest to the top left corner criteria (0 to 0.05) at a threshold of 1.047 specific gravity. For all RID IgG thresholds, the best Brix threshold (regardless of type or whether the sample was fresh or frozen) was 18 or 19%, with the highest Youden indices (range: 0.47 to 0.61) and lowest distance to the top left corner criteria (range: 0.09 to 0.16); however, we recommend 19%, because this reduces the potential of feeding poor-quality colostrum. The ELISA method was the poorest predictor of colostrum concentration. Age was not found to affect colostrum quality; however, the sample size of this subset was small. Hydrometers are inexpensive and easy to use, whereas Brix methods use only a small amount of colostrum; we suggest that either method could be used on-farm.