A meta-analysis of the effects of feeding yeast culture produced by anaerobic fermentation of Saccharomyces cerevisiae on milk production of lactating dairy cows

The purpose of this study was to use meta-analytic methods to estimate the effect of a commercially available yeast culture product on milk production and other production measures in lactating dairy cows using a meta-analysis of randomized controlled trials. Sixty-one research publications (published journal articles, published abstracts, and technical reports) were identified through a review of literature provided by the manufacturer and a search of published literature using 6 search engines. Thirty-six separate studies with 69 comparisons met the criteria for inclusion in the meta-analysis. The fixed-effect meta-analysis showed substantial heterogeneity for milk yield, energy-corrected milk, 3.5% fat-corrected milk, milk fat yield, and milk protein yield. Sub-group analysis of the data showed much less heterogeneity in peer-reviewed studies versus non-peer-reviewed abstracts and technical reports, and tended to show higher, but not significantly different, treatment effects. A random-effects meta-analysis showed estimated raw mean differences between treated and untreated cattle reported in peer-reviewed publications of 1.18 kg/d [95% confidence interval (CI): 0.55 to 1.81], 1.61 kg/d (95% CI: 0.92 to 2.29), and 1.65 kg/d (95% CI: 0.97 to 2.34) for milk yield, 3.5% fat-corrected milk, and energy-corrected milk, respectively. Milk fat yield and milk protein yield for peer-reviewed studies showed an increase in the raw mean difference of 0.06 kg/d (95% CI: 0.01 to 0.10) and 0.03 kg/d (95% CI: 0.00 to 0.05), respectively. Estimated raw mean dry matter intake of the peer-reviewed studies during early lactation (<70 d in milk) and not-early lactation were 0.62 kg/d (95% CI: 0.21 to 1.02) and a decrease of 0.78 kg/d (95% CI: -1.36 to -0.21), respectively. These findings provide strong evidence that this commercially available yeast culture product provides significant improvement in several important milk production outcomes as evaluated in production settings typical for commercial dairies in North America. Utilizing meta-analytic methods to study the complete breadth of information relating to a specific treatment by studying multiple overcomes of all eligible studies can reduce the uncertainty often seen in small individual studies designed without sufficient power to detect differences in treatments.