Abstract We conducted a meta-analysis focused on the effects of dietary total sulfur (TS) and adjusted ruminal protein sulfur (ARPS) on beef cattle growth performance, carcass characteristics, apparent total tract digestibility, and ruminal fermentation. Peer-reviewed articles and dissertations/thesis from Google Scholar, PubMed, and Web of Science databases (n = 110) were used in a 5-step PRISMA evaluation process, resulting in a final sample of 22 articles that included 37 comparisons with 8,842 observations. The intervention studies included the comparison of high sulfur (HS = 0.52 and 0.40% DM basis, treatment group mean) vs. low sulfur (LS = 0.23 to 0.15% DM basis, control group mean), with biological responses measured for both dietary TS and ARPS, respectively. Meta-regression analysis was performed with studies that included three levels of ARPS (averages for low= 0.14%, medium = 0.25%, and high = 0.33% levels). The ARPS, which reflects S available within the rumen, was calculated from published equations, with tabular feed composition data used to acquire feed ingredient protein degradability coefficients. Data were analyzed with a mixed-model approach using R software and Metafor package (version 4.2). Forest and funnel plots were created using STATA software and used to evaluate the potential for publication bias in the data set. The I2 statistics and Q value (reflecting heterogeneity) were calculated and considered inconsistent when I2 50% and Q ≤ 0.05. Meta-regression analysis data were analyzed with a mixed-model approach that included random slopes and intercepts for studies. Independent of the sources of S offered in the diet, greater (P ≤ 0.01) S intake (g/d) was observed when HS was offered, expressed either as total S or ARPS. The HS diets, assessed as ARPS, increased (P 0.01) ruminal hydrogen sulfide (mMol/L) and decreased (P = 0.04) propionate mMol/L, whereas no effect (P = 0.19) was observed for apparent total tract DM digestibility, ruminal pH (P = 0.45), total VFA (m/L, P = 0.69), acetate (mMol/L, P = 0.92), and acetate: propionate ratio (P = 0.48). The HS diets evaluated for both total S and ARPS decreased DMI (P ≤ 0.01), ADG (P ≤ 0.01), marbling score (P ≤ 0.03), and yield grade (P ≤ 0.04), but no effect was observed for G: F (P ≥ 0.58). Nevertheless, the HS diets assessed with ARPS decreased the hot carcass weight (HCW, P = 0.01), dressing percent (P = 0.04), longissimus muscle area (P = 0.04), and 12th-rib fat (P = 0.01), whereas the evaluation based on total S did not detect (P ≥ 0.18) these differences, except a tendency (P = 0.07) for a detrimental effect on HCW. In the meta-regression analyses, quadratic effects (P ≤ 0.01) were found for DMI, ADG, G:F, and HCW based on ARPS level. Statistical evidence of the negative effects of high S was identified for most of the variables when evaluated via ARPS rather than dietary total S. Thus, we conclude that the dietary ARPS is a better indicator than dietary total S to predict detrimental effects of dietary excess S when formulating diets for beef cattle, regardless of the S source used.
Melgar et al. (Wed,) studied this question.