Objective: To develop a new risk score as a predictive tool from clinical risk factors for determining high-risk pregnancy with
shoulder dystocia.
Materials and Methods: A retrospective study was performed. The demographic data, prenatal care history, and the risk factors, such as total weight gain, estimated fetal weight, and the instrumental assisted vaginal delivery (InVD), were recorded and compared between the shoulder dystocia (SD) and non-shoulder dystocia (non-SD) groups. The risk score for prediction was developed by multivariate logistic regression analysis.
Results: Of 872 vaginal deliveries, 42 SD cases were collected and 830 non-SD cases were included. In a multivariate analysis, there were three clinical risk factors, statistically significant; total weight gain >16 kg (TWG16), Estimated fetal weight >3,200 grams (EFW3200) and the InVD. The odd ratios of these risk factors were calculated and converted to the risk score. The final score model to predict shoulder dystocia had receiver operating characteristic curve of 79.73%. Each patient was given a score as the presented risks; TWG16 = 2, EFW3200 = 3, and InVD = 5. Then, the summation of the score was divided into low-, intermediate-, and high-risk groups at cut off value scores of 0 to 4, 5 to 6, and >7, respectively. Positive likelihood ratio in those groups were 0.12, 5.94 and 10.97, in orderly, with statistically significance (p-value <0.001).
Conclusion: The study has developed an easy and practical new risk score for predicting pregnant women who at risk of shoulder dystocia.

Keywords: Predictive score, Shoulder dystocia