Relationships between physically effective undegraded neutral detergent fiber and rumen-fermentable starch in diets for lactating Holstein dairy cows.

We assessed the effect of rumen-fermentable starch (RFS; 7-h in vitro starch digestibility) in diets that contained low or moderate physically effective undegraded NDF after 240-h of in vitro fermentation (peuNDF240) on DMI, milk yield and composition, chewing behavior, ruminal function, and apparent total-tract nutrient digestibility. Sixteen lactating Holstein cows (8 ruminally cannulated) averaging 85 ± 15 (SD) DIM were enrolled in a replicated 4 × 4 Latin square design (28-d periods) with a 2 × 2 factorial arrangement of treatments. Diets differed in peuNDF240 and RFS content by inclusion of either brown midrib or conventional corn silage and by varying relative proportions of ground corn and beet pulp. The peuNDF240 was measured as undegraded NDF after 240 h of in vitro fermentation (uNDF240) content of the physically effective fraction of the diet (particles retained on ≥1.18-mm screen with dry vertical sieving). All diets contained low uNDF240 (∼7% of DM), but its distribution was skewed to coarser particles for the moderate peuNDF240 diets. The Analyzed composition of diets was as follows: (1) low peuNDF240 (6.4% of DM), low RFS (16.7% of DM); (2) low peuNDF240 (6.1% of DM), moderate RFS (19.2% of DM); (3) moderate peuNDF240 (8.6% of DM), low RFS (16.9% of DM); and (4) moderate peuNDF240 (8.0% of DM), moderate RFS (19.0% of DM). Cows were housed in tiestalls, fed TMR once daily, and milked 3 times daily. Data were analyzed as a factorial arrangement of treatments within a replicated Latin square design using the MIXED procedure of SAS (version 9.4, SAS Institute Inc., Cary, NC). Model included fixed effects of RFS, peuNDF240, RFS × peuNDF240, period within replicated square, and square. Cow within square was tested as a random effect. Few interactions existed between peuNDF240 and RFS for measures of intake, lactation performance, and ruminal characteristics. Cows fed low peuNDF240 diets consumed more DM than cows fed moderate peuNDF240. As expected, moderate peuNDF240 diets resulted in greater uNDF240 intake, whereas moderate RFS diets boosted starch intake. Cows fed diets containing moderate RFS produced less milk fat and 3.5% FCM than cows fed low RFS. Low-peuNDF240 diets resulted in fewer mixed origin fatty acids in milk fat and greater degree of unsaturation. Relatedly, moderate-RFS diets tended to reduce efficiency of FCM production (FCM/DMI) compared with low RFS. Eating time was unaffected by diet, but with low peuNDF240, diets of moderate RFS boosted rumination activity. Additionally, moderate-RFS diets reduced meal length and increased daily meal bouts. Consistent with changes in milk fat, low-RFS diets increased acetate:propionate ratio, whereas low-peuNDF240 diets reduced molar proportions of branched-chain VFA. Diet had no effect on measures of ruminal pH. As expected, moderate dietary RFS tended to increase ruminal pool size of starch, and moderate peuNDF240 increased pool size of uNDF240, although there were few dietary effects on ruminal turnover of starch or uNDF240. Low-peuNDF240 diets resulted in greater apparent total-tract NDF digestibility than moderate peuNDF240. Feeding even moderate dietary RFS depresses milk fat output when low-peuNDF240 diets are fed to lactating dairy cows.