A multiple regression model approach to contrast the performance of 2-hydroxy-4-methylthio butanoic acid and DL-methionine supplementation tested in broiler experiments and reported in the literature.

The purpose of this paper was to compile all available literature comparing the relative performance of 2-hydroxy-4-methylthio butanoic acid (HMTBA) with DL-methionine (DLM) in broiler chickens and using multiple regression techniques, to estimate the predicted dose responses and relative performance of the 2 Met compounds for gain and feed conversion (FC). A database was developed that contained all available broiler studies in which HMTBA and DLM were both present in the same study; weight gain was recorded; Met addition, age of birds, and duration of study were defined; and an unsupplemented control treatment was present. Sixty-two references complied with these criteria and included 100 experiments with 427 observations for HMTBA and 411 for DLM. Multiple regression analysis of the database was used to identify the experimental and dietary conditions that contributed to the gain and FC responses of each source of Met activity. All identified variables contributed similarly to each Met source prediction model and both gain and FC models described a quadratic dose response. Under the average conditions of the database, the predicted responses for gain and FC models did not significantly differ between HMTBA and DLM. However, a trend was observed (P < or = 0.1) for the peak gain response for HMTBA to be numerically greater than DLM, suggesting benefits of HMTBA over DLM in the region of supplementation that is commercially relevant. The experimental and nutritional conditions that contribute to the response to HTMBA and DLM were identified and are discussed in the paper. This statistical approach provided a means to summarize the results obtained from a multitude of studies conducted over the last 5 decades and has provided a meaningful estimate of the relative performance of the 2 sources of Met activity. The lack of differences between the 2 predicted models under experimental and commercial conditions supports an overall conclusion of equal performance of DLM and HTMBA when compared on an equal molar basis.