Methionine is considered a limiting AA for milk protein synthesis in dairy cows, and supplementation of unprotected, free Met is nearly 100% degraded by ruminal microorganisms. 2-hydroxy-4-methylthio-butanoic acid (HMTBa) can be converted to Met in the body, and has been used as a Met alternative in dairy production. However, results of published studies assessing the effects of HMTBa supplementation on performance variables are inconsistent. A meta-analysis was performed to assess relationships between HMTBa supplementation and nutrient digestibility and milk performance. Data for HMTBa dose, dietary information and major performance variables (rumen VFA composition, milk performance, nutrient digestibility) were collected from 48 articles resulting in 204 treatment means. The HMTBa supplementation effects on response variables were analyzed using a linear mixed model that included random study effects when significant. Otherwise, a simple linear model was used. To HMTBa supplementation was tested both as a numerical (Met equivalent dosage) and categorical (fed/not fed) predictor. Other explanatory variables included in the models were study type (in vitro or in vivo), forage percent, NDF percent, CP percent and days in milk (DIM). Results showed that HMTBa supplementation increased microbial N output (P < 0.001), blood Met concentration (P = 0.007), milk fat yield (P < 0.05), and DM (P = 0.09), OM (P < 0.001) and CP (P = 0.009) digestibility. Although HMTBa supplementation increased milk fat yield, further studies are needed to elucidate the mechanisms by which HMTBa generates this response. To further evaluate how HMTBa supplementation should be accounted for in nutrient requirement models like the NRC (2001), standardized residuals of digested NDF and CP, and microbial N (observed minus predicted as a percentage of predicted) were regressed against Met equivalent of HMTBa to identify adjustments so as to better model the effects of HMTBa supplementation on nutrient supply. The predicted amount of NDF and CP digested was found to increase by 1.08% and 1.36% of predicted digested NDF and CP, respectively for each g of Met equivalent of HMTBa added to the diet.