Several in vivo CH measurement techniques have been developed but are not suitable for precise and accurate large-scale measurements; hence, proxies for CH emissions in dairy cattle have been proposed, including the milk fatty acid (MFA) profile. The aim of the present study was to determine whether recently developed MFA-based prediction equations for CH emission are applicable to dairy cows with different diacylglycerol o-acyltransferase 1 (DGAT1) K232A polymorphism and fed diets with and without linseed oil. Data from a crossover design experiment were used, encompassing 2 dietary treatments (i.e., a control diet and a linseed oil diet, with a difference in dietary fat content of 22 g/kg of dry matter) and 24 lactating Holstein-Friesian cows (i.e., 12 cows with DGAT1 KK genotype and 12 cows with DGAT1 AA genotype). Enteric CH production was measured in climate respiration chambers and the MFA profile was analyzed using gas chromatography. Observed CH emissions were compared with CH emissions predicted by previously developed MFA-based CH prediction equations. The results indicate that different types of diets (i.e., with or without linseed oil), but not the DGAT1 K232A polymorphism, affect the ability of previously derived prediction equations to predict CH emission. However, the concordance correlation coefficient was smaller than or equal to 0.30 for both dietary treatments separately, both DGAT1 genotypes separately, and the complete data set. We therefore concluded that previously derived MFA-based CH prediction equations can neither accurately nor precisely predict CH emissions of dairy cows managed under strategies differing from those under which the original prediction equations were developed.