Avaliação de modelos empíricos para determinação da taxa de passagem de caprinos em crescimento

• Ano de defesa: 2015
• Autor(a) principal: Bompadre, Thiago Francisco Ventoso [UNESP]
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Estadual Paulista (Unesp)
• Palavras-chave em Português: Caprino; Caprino - Nutrição; Digestão; Rúmen; Ruminante
• Link de acesso: http://hdl.handle.net/11449/136108; http://www.athena.biblioteca.unesp.br/exlibris/bd/cathedra/15-02-2016/000858151.pdf

The passage rate (kp) is related to several factors, such as intake, digestion and nutrient absorption. Fiber digestion is greater in the reticulorumen segment. Considering the importance of kp in ruminant nutrition, some studies have been developing models to predict kp. In the literature, there are many models to predict kp in ruminants, but few models consider exclusively goats. Thus, this study aimed to evaluate most widely used models to determine kp in growing goats. The database used to evaluate the literature models derived from two studies with Saanen goats of different growth stages between 15 and 45 kg of body weight. A total of 102 goats fed ad libitum were used. The studies used the ruminal evacuation technique in slaughtered animals. The markers used were Ytterbium (external marker of solid phase) and indigestible NDF (internal marker of solid phase). Five kp models published between 1960 and 2013 were evaluated. The evaluation of the literature models was conducted using the regression of residual (observed minus predicted) on centered predicted values. Considering the model evaluation for Ytterbium in the observed data, none of the models tested fit well in this database. On the other hand, when the models were evaluated using the indigestible NDF database, the model developed by Krizsan et al. (2010) was accurate (mean bias, P = 0.82) and did not present systematic error in the prediction variance (linear bias, P = 0.45). A covariance analysis were performed for identifying variables to improve the adequacy of the models. The theoretical model developed by Vieira et al (2008) after integrating iNDF intake, showed mean bias (P = 0.42) and linear bias (P = 0.75). In conclusion, in ruminal evacuation technique studies using iNDF as marker, the models developed by Vieira et al. (2008) adjusted by iNDF intake and Krizsan et al. (2010) could provide a reliable prediction of kp for growing goats