Noriko Furuita1,2, Satoe Okabayashi3, Ayumi Yamanishi2, Misako Nakamura2, Takashi Kawamura4. 1. Department of Preventive Services, School of Public Health in the Graduate School of Medicine, Kyoto University, Yoshida-honmachi, Sakyo-ku, Kyoto, 606-8501, Japan. 2. Kyoto Min-iren Chuo Hospital, 2-1 Uzumasa Tsuchimoto-cho, Ukyo-ku, Kyoto, 616-8147, Japan. 3. Kyoto University Health Service, Yoshida-honmachi, Sakyo-ku, Kyoto, 606-8501, Japan. okabayashi.satoe.8c@kyoto-u.ac.jp. 4. Kyoto University Health Service, Yoshida-honmachi, Sakyo-ku, Kyoto, 606-8501, Japan.
Abstract
PURPOSE: An emergency caesarean section (CS) has more complications than a planned CS. The arrest of labour is a major indication for an emergency CS. This study aimed to develop a prediction model for the arrest of labour to be used in regular check-ups at 36 or 37 gestational weeks for primiparas. METHODS: This was a retrospective cohort study conducted at a single institution in Japan using data from January 2007 to December 2013. Primiparas attending regular check-ups during 36 or 37 gestational weeks, with live single foetuses in a cephalic presentation were included. The outcome was the incidence of labour arrest. Candidate predictors included 25 maternal and foetal findings. We developed a prediction model using logistic regression analysis with stepwise selection. A score was assigned to each predictor of the final model based on their respective β coefficients. RESULTS: A total of 739 women were included in the analysis. Arrest of labour was diagnosed in 47 women (6.4%), and all of them delivered by emergency CS. The predictors in the final model were a Bishop score ≤ 1, maternal height ≤ 154 cm, foetal biparietal diameter ≥ 91 mm, pre-pregnancy weight ≥ 54 kg, maternal haemoglobin concentration ≥ 11.0 g/dl, and amniotic fluid index ≥ 13. The area under the receiver operating characteristic curve was 0.783. CONCLUSION: We have developed the first model to predict arrested labour before its onset. Although this model requires validation using external samples, it will help clinicians and pregnant women to control gestational conditions and make decisions regarding planned CS.
PURPOSE: An emergency caesarean section (CS) has more complications than a planned CS. The arrest of labour is a major indication for an emergency CS. This study aimed to develop a prediction model for the arrest of labour to be used in regular check-ups at 36 or 37 gestational weeks for primiparas. METHODS: This was a retrospective cohort study conducted at a single institution in Japan using data from January 2007 to December 2013. Primiparas attending regular check-ups during 36 or 37 gestational weeks, with live single foetuses in a cephalic presentation were included. The outcome was the incidence of labour arrest. Candidate predictors included 25 maternal and foetal findings. We developed a prediction model using logistic regression analysis with stepwise selection. A score was assigned to each predictor of the final model based on their respective β coefficients. RESULTS: A total of 739 women were included in the analysis. Arrest of labour was diagnosed in 47 women (6.4%), and all of them delivered by emergency CS. The predictors in the final model were a Bishop score ≤ 1, maternal height ≤ 154 cm, foetal biparietal diameter ≥ 91 mm, pre-pregnancy weight ≥ 54 kg, maternal haemoglobin concentration ≥ 11.0 g/dl, and amniotic fluid index ≥ 13. The area under the receiver operating characteristic curve was 0.783. CONCLUSION: We have developed the first model to predict arrested labour before its onset. Although this model requires validation using external samples, it will help clinicians and pregnant women to control gestational conditions and make decisions regarding planned CS.