Abstract

The aim of this study was to investigate factors affecting, and estimate variance components of, stillbirth using data collected as part of Livestock Improvement’s Sire Proving Scheme. Only calves that reached full-term of gestation were considered (i.e. abortions and inductions were excluded). After editing there were 773,904 records on calf-fate collected from 1987 to 2004. The mean incidence of stillbirth was 7.2%. The fixed effect model fitted to the data included the effects of contemporary group, birth assistance code, the sex of the calf, the gestation length of the dam, whether the dam was primiparous or multiparous and mating type e.g. a Jersey cow mated to a Holstein-Friesian sire etc. Breed proportions of New Zealand (NZHF) and North American (OSHF) origin Holstein-Friesian and Jersey were also fitted in addition to the proportion of heterosis expected in the calf. Data were analysed by considering stillbirth as a binomial trait using the logit transformation in ASReml that included a sire-maternal-grandsire pedigree. The direct heritability of stillbirth was estimated to be 0.010 (s.e. 0.009). It was not possible to estimate the maternal heritability using this dataset. Estimates of fixed effects were obtained from ASReml by analysing stillbirth as a continuous trait, as back-transformed fixed effect estimates obtained using a binomial model are unreliable when the incidence is small. The largest effect on the probability of stillbirth was calving difficulty. Compared to no assistance at calving, the two categories of greatest calving difficulty resulted in 22% and 41% mortality. The next largest effect on stillbirth was gestation length, with longer gestations being associated with more stillbirths. Males were 1% more likely to be stillborn than females. OSHF were most likely to be stillborn, followed by Jerseys and then NZHF. Heterosis effects were significant and favourable, except between the two strains of Holstein-Friesians where there was no significant heterosis. The largest heterosis estimate was between NZHF and Jerseys. Unlike some overseas programmes, genetically selecting for reduced stillbirths would not be effective in New Zealand, crossbreeding and devising strategies to manage calving assistance and gestation length appear to be more worthwhile.

JE, Pryce, BL Harris, S Sim, and AW McPherson

Proceedings of the New Zealand Society of Animal Production, Volume 66, Napier, 98-102, 2006