RESUMEN
OBJECTIVE: Our study aims to identify risks factors associated with failed trial of scar and to set a predictive score of labour in women with scarred uterus in the Democratic Republic of the Congo. METHODS: We conducted a multicenter cross-sectional analytic study of patients with scarred uterus in four maternity units in the Democratic Republic of the Congo (DRC) from 1 January to 31 December 2013. Logistic regression model was used to identify factors associated with failed trial of scar. We set a predictive score based on this model in order to predict trial of scar failure in maternity units in the DRC. ROC curve was used to assess the ability of the set score to identify patients at risk for trial of scar failure. The cut off point for the predictive score was determined on the basis of the Youden-index-based optimal sensitivity and specificity. All the tests in our study were carried out by using a significance threshold of α=0.05. RESULTS: Two explanatory factors in trial of scar failure were retained. They were the socio-demographic factor (maternal age) and three obstetric factors (fundal height, fetal presentation and premature rupture of membranes). Predictive score was set to predict trial of scar failure. This score was determined based on four elements: maternal age, evaluation of the gestational sac on admission, fundal height and fetal presentation. The minimum score was set at 4 and the maximum score was set at 16. The threshold value was set at 7. A total score greater than or equal to 7 reflects a risk of trial of scar failure. CONCLUSION: We set a predictive score to predict trial of scar failure. This score was determined based on four elements: maternal age, evaluation of the gestational sac on admission, fundal height and fetal presentation. A total score greater than or equal to 7 reflects a risk of trial of scar failure.