Nomogram-based prediction of placental abruption in severe pre-eclampsia based on serum APN, Cys-C, and D-dimer.

Objective: This study aimed to construct a nomogram model for predicting placental abruption in patients with severe pre-eclampsia based on serum adiponectin (APN), cystatin C (Cys-C), and D-dimer, and to validate its predictive efficacy and clinical application value. Methods: A total of 256 patients with severe pre-eclampsia who were treated in our hospital from December 2021 to January 2025 were enrolled in this retrospective study. They were divided into a training set (n = 179) and a validation set (n = 77) using the random number table method. General information, clinical indicators, and serum levels of APN, Cys-C, and D-dimer of the patients were collected. In the training set, risk factors for placental abruption were screened through univariate analysis and multivariate logistic regression analysis, and a nomogram prediction model was constructed. The predictive efficacy of the model was evaluated by the receiver operating characteristic curve (ROC) and calibration curve, and then validated in the validation set. The clinical application value of the model was evaluated by decision curve analysis (DCA). Results: In the training set, 44 cases (24.93%) had placental abruption, while in the validation set, 19 cases (25.06%) did. There were no statistically significant differences in the incidence of placental abruption and clinical characteristics between the two groups (p > 0.05). Multivariate logistic regression analysis showed that decreased serum APN level, increased Cys-C and D-dimer levels, proteinuria quantification during pregnancy ≥5 g/24 h, and oligohydramnios were independent risk factors for placental abruption in patients with severe pre-eclampsia (all p < 0.05). The C-index of the constructed nomogram model in the training set and validation set was 0.809 and 0.730, respectively. The ROC curve showed that the area under the curves of the model for predicting placental abruption in the training set and validation set was 0.809 (95% CI: 0.722-0.896) and 0.730 (95% CI: 0.492-0.969), respectively, and the sensitivities and specificities were 0.588, 0.924, and 0.600, 0.840, respectively. Conclusion: The nomogram model constructed based on serum APN, Cys-C, and D-dimer has good predictive efficacy for placental abruption in patients with severe pre-eclampsia, which is helpful for the early prediction of the risk of placental abruption, guiding clinical decision-making, and ensuring the safety of mothers and infants.