The esterification of 2-hydroxy-4-(methylthio)-butanoic acid (HMB) to isopropanol (HMBi) decreases the rate and extent of its ruminal breakdown. The modes of action of HMB and HMBi appear to be different. The quantification of the production response to HMBi has not been done. The objectives of this study were (1) to determine the lactation response to HMB, (2) to determine the lactation response to HMBi, and (3) to evaluate whether the response to HMBi is affected by HMB in the diet. Sixty-one Holstein cows (24 primiparous, 37 multiparous) were assigned to 1 of 4 dietary treatments 21 to 28 d after calving. The base diet consisted of [on a dry matter (DM) basis] 32.5% corn silage, 17.5% alfalfa hay, 10% whole cottonseed, and 40% of a pelleted concentrate made primarily of ground corn, soybean meal, and blood meal, and was fed for 16 wk as a control diet. To prepare the dietary treatments, the base diet was supplemented with 0.1% of diet DM with HMB (treatment 2), with 0.15% HMBi (treatment 3), or with 0.045% HMB and 0.15% HMBi (treatment 4). Results showed a significant increase in milk yield (2.9 kg/d), protein content (0.15%), protein yield (115 g/d), fat yield (165 g/d), and lactose yield (182 g/d) from HMBi. Supplementation of HMB had small and nonsignificant effects on milk yield and composition. There were no significant interaction effects of HMB with HMBi on any of the production traits measured in this experiment. Plasma free Met as a proportion of essential amino acids was increased by HMBi, but not by HMB. Dietary supplementation of HMBi increased gross N efficiency expressed as the proportion of ingested N secreted in milk. Consequently, HMBi significantly improved N efficiency.