DEVELOPMENT AND INTERNAL VALIDATION OF A PREDICTION MODEL TO PREDICT THE PROBABILITY OF FUTURE RECURRENCE OF HYPERTENSION IN PATIENTS WITH HYPERTENSIVE DISORDERS OF PREGNANCY

Objective: This study aimed to obtain a clinical prediction model for the future recurrence of hypertension in women with a history of gestational hypertension or Preeclampsia-Eclampsia by analyzing the basic data and laboratory indicators in their pregnancy. Design and method: We retrieved hypertensive disorders of pregnancy (HDP) patients of the First Affiliated Hospital of Dalian Medical University and the Second Affiliated Hospital of Dalian Medical University from 2001 to June 2021. After telephone follow-up, a total of 605 pregnant women with previous HDP were followed up. The baseline data of the cases were followed by binary logistic regression analysis according to the follow-up hypertension outcome, Nomogram plotted according to independent risk factors, and the model discrimination was tested by operating characteristic curve (ROC), The reproducibility of the model was verified by internal validation on the development dataset (Bootstrap self-sampling method, internal cross-validation and Jackknife validation), verify the calibration of the model through the calibration curve (Hosmer-Lemeshow goodness of fit test), and Validation of clinical predictive value by Decision Curve Analysis (DCA). Results: data analysis showed that A total of 200 (33.06%) pregnant women with a previous history of HDP had recurrent hypertension during follow-up (median follow-up time was 5 years). The predictors in the Nomogram were the highest diastolic blood pressure (DBP) during pregnancy, the highest systolic blood pressure (DBP) during pregnancy, peak Alkaline phosphatase value, age at delivery, peak uric acid value, and positive urine protein, Contrary to the remaining 5 predictors, a low level of peak alkaline phosphatase value predicted a high risk of postpartum hypertension. the Area Under Curve (AUC) of ROC about the nomogram was 0.797, and a similar C-index was obtained in the internal validation, and this model show a high fitting degree in the calibration curve and show clinical predictive value. Conclusions: The results of this study can provide predictive indicators for women with HDP to suffer from hypertension after delivery, which can be used for clinical decision-making, and aid with follow-up blood pressure monitoring and the selection of primary prevention treatment groups.