Risk factors for and reproductive outcomes of phantom cows on New Zealand dairy farms

AIMS: To determine some of the risk factors for cows not observed in oestrus within 35-42 days of an unsuccessful artificial insemination (AI; phantom cows), and the reproductive outcomes and effect of treatment of phantom cows. MATERIALS AND METHODS: Over 2 years, in dairy herds from the Waikato (n=10) and Canterbury (n=4) regions of New Zealand, pregnancy diagnosis was carried out 35-42 days after AI on cows that had been inseminated in the first 3 weeks after the start of mating (PSM) but had not been seen returning to oestrus. Risk factors for phantom cows were analysed using a generalised linear mixed effect model. In Year 1, all phantom cows were left untreated. In Year 2, phantom cows were categorised as having a corpus luteum (CL) (CL+ n=120), or having ovarian follicles >= 10 (n=101) or <10 (n=40) mm in diameter. Cows with a CL were treated with cloprostenol or untreated and placed with bulls. Cows with no CL received intravaginal progesterone (P4) for 7 days, with injection of gonadotrophin-releasing hormone (GnRH) on Days 0 and 9, and cloprostenol on Day 7 followed by AI. Pregnancy diagnosis of all cows took place 100-120 days after PSM and interval to conception and final pregnancy rate determined. RESULTS: Overall, of cows inseminated in the first 3 weeks after PSM that did not return to oestrus, 610/6,734 (9.1%) were phantom cows. From the final multivariable analysis, treatment for anoestrus, BCS <= 4.0 at mating, being 2 or >6 years of age, and pure-bred, and decreasing interval between calving and mating, until 98 days post calving, were associated with increased odds of being a phantom cow. Compared to all other groups of cows, phantom cows had a longer interval to conception (p<0.001) and a lower final pregnancy rate (p<0.001). Treatment of CL+ cows or cows with follicles >= 10 mm did not affect reproductive outcomes (p>0.3). For cows with follicles <10 mm treatment decreased the final percentage not pregnant (3/27; 11%; p=0.01) and interval to conception (21 days; p=0.02) compared with controls (7/13; 54% and 37 days, respectively). CONCLUSIONS AND CLINICAL RELEVANCE: Risk factors for phantom cows were identified that could be manipulated to reduce the number of phantom cows in a herd, in particular increasing BCS. Treatment of the majority of phantom cows did not improve reproductive performance.