A comparison of in situ and in vitro methods to estimate in vivo fermentable organic matter of forages in ruminants

Several known in situ and in vitro methods were compared for their reliability for determining directly or indirectly in vivo fermentable organic matter (in vivo FOM) of forages in ruminants. Twelve forage types were used: fresh and conserved forms of lucerne, red clover, orchard grass and perennial ryegrass. Organic matter truly digested in the rumen which in our study was regarded as equivalent to in vivo FOM was determined in six cannulated sheep, using the flow markers %sup51;Cr-EDTA and &sup103;Ru-Phenanthrolin. In vivo FOM was estimated directly from results of the in situ nylon bag technique using three cows, and from the results of three in vitro methods, and indirectly by calculating in vivo FOM using equations from the Dutch and French protein evaluation systems. The in vitro methods were an enzymatic technique using pepsin and cellulase, the method of Tilley & Terry and the gas production technique. In vivo FOM was best correlated (R² = 0.74; n = 12) with gas production after 20 hours of incubation. The correlation improved when fresh and conserved forages were considered separately (R² = 0.90; n = 12). Indirectly, in vivo FOM was well estimated from the results of the in situ, the gas production and the Tilley & Terry methods (R² = 0.760.80; n = 12). The accuracy of the direct and indirect in vivo FOM estimates was similar. However, the direct in vivo FOM estimate was a regression and the indirect estimate was a validation. In conclusion, in vivo FOM was best estimated indirectly using the equation from the Dutch protein evaluation system, whereas the estimate was more accurate with the in situ and the gas production techniques than when the other in vitro methods were used