Endometritis Moges & Jebar, 2012) and 4 to

Endometritis is a postpartum disease characterized by an inflammation of the uterine endometrium 20-68 days postpartum period (dpp) and without signs of illness (Chun-Jie et al., 2013; Pascottini et al., 2017).The disease may be clinical endometritis (CLE) or subclinical endometritis (SCLE). The CLE is characterized by presence of purulent or mucopurulent uterine discharge detectable in the vagina 20 to 60 dpp (Chun-Jie et al., 2013; Tayebwa et al., 2015). In contrast, the SCLE is characterized by abnormal presence of the proportion of polymorphonuclear neutrophils (PMN) cells in endometrial cytology samples collected between 20 to 68 dpp (Kasimanickam et al., 2005; Pascottini et al., 2017). Both CLE and SCLE are prevalent in dairy cows and disrupts reproductive cycles in positive cows resulting in reduced performance and profitability in the herd (Sheldon et al., 2009; Dubuc et al., 2010). The estimated prevalence rates of endometritis suggests large variability, 11 to 67% for SCLE (Ahmadi et al., 2016; Moges & Jebar, 2012) and 4 to 43% for CLE (Leutert et al., 2012; Tayebwa et al., 2015), depending on production systems and breed, which suggests that there are opportunities to manage the disease with good understanding of the risks involved. Reports on prevalence of CLE and SCLE are available for industrial dairy systems of USA, Asia and Europe where dairy herds are managed in confinement in-group housing units in which mating is predominantly with Artificial Insemination (AI) services. In Rwanda, 92% of the smallholder dairy herds are managed under confinement as well in zero-grazing housing units and served with AI services (Nishimwe et al., 2015; Mutimura et al., 2013). The AI success rate is estimated to be below 35% (Makuza et al.., 2016; Mwabonimana & Habimana, 2015). This is evidence of sub optimal fertility which is evidenced by long days-open (191.5±70.6), high number of services per conception (3 ± 1.3) and long calving intervals (17.22± 6.28 months) and anoestrus postpartum of 44% attained in the Rwandan dairy herds (Bishop & Pfeiffer, 2008; Makuza et al.,., 2016). The Ankole indigenous cattle, crossbred and exotic breeds all exhibit sub optimal fertility in interval calving  first estrus, respectively, 8.78±7.66; 8.12±8.89 and 8.7±7.8 months, which can be associated with resulting sub optimal milk production estimated at 2.1 litres/day for Ankole, 5.5 litres/day for crossbred, and 10litres/day for exotic breeds (MINAGRI, 2015). The low AI success rate below 35% and sub optimal fertility and production observed in the Rwandan dairy herds has been explained as resulting from poor management but without identifying the specific risks factors involved. One likely area of management failure could be in managing uterine health, which is essential for success of AI services. Cow uterine that is not healthy disrupts the uterine homeostasis and results in conception failures (Galvão, 2011; Hajurka et al., , 2005). It is therefore possible that smallholder zero-grazed cows being AI served could be at risk of prevalent endometritis disease. This is a prevalent uterine disease in cows, but in Rwanda to date, empirical evidence is yet to be presented to support the presence or absence of endometritis disease. In dairy herds, several studies have associated presence of endometritis to significant production losses. The production losses arise from unproductive days in the herds, abortion, mortality and morbidity (Ali, 2011; Cattaneo et al., 2015). The risk factors associated with endometritis prevalence are those of cow and farm level sources.  The main risk at the cow level include: parity, body condition score, multiple gestation, dystocia, metabolic disorders, age, breed, calf sex, retained placenta, stillbirth, gestation length, mastitis and brucellosis (Cheong et al., 2011; Giuliodori et al., 2013; Mushonga et al., 2017). The potential risk factor at farm level include calving environment, calving season, housing in periparturient period, herd size, calving pen material, calving environment, feeding system and breeding services ( Moges & Jebar, 2012; Tayebwa et al., 2015). These risks are relevant in the smallholder herds and there could predispose cows to endometritis but empirical evidence are lacking as to which ones of them pose high risks for CLE or SCLE. Such evidence would be valuable in informing management interventions to target the high risk factors.