RESUMEN
PURPOSE: To describe the association between quantity of blood loss, duration of the third stage of labour, retained placenta and other risk factors, and to describe the role of a retained placenta depending on the cutoff used to define postpartum haemorrhage. METHODS: Cohort study of all vaginal deliveries at two Danish maternity units between 1 January 2009 and 31 December 2013 (n = 43,357), univariate and multivariate linear regression statistical analyses. RESULTS: A retained placenta was shown to be a strong predictor of quantity of blood loss and duration of the third stage of labour a weak predictor of quantity of blood loss. The predictive power of the third stage of labour was further reduced in the multivariate analysis when including retained placenta in the model. There was an increase in the role of a retained placenta depending on the cutoff used to define postpartum haemorrhage, increasing from 12% in cases of blood loss ≥ 500 ml to 53% in cases of blood loss ≥ 2000 ml CONCLUSION: The predictive power of duration of the third stage of labour in regard to postpartum blood loss was diminished by the influence of a retained placenta. A retained placenta was, furthermore, present in the majority of most severe cases.