Objective To investigate the risk factors of gestational diabetes mellitus (GDM) combined with gestational hypertension, and to establish a nomogram prediction model.
Methods According to the ratio of 2 to 1, 294 GDM patients were divided into modeling group (196 cases) and verification group (98 cases). The modeling group was divided into the hypertensive group (51 cases) and the non-hypertensive group (145 cases). The influential factors of GDM combined with pregnancy-induced hypertension were analyzed, and the nomogram model was established and verified.
Results Pre-pregnancy overweight/obesity, excessive weight gain during pregnancy, three abnormalities in oral glucose tolerance test, homocysteine (Hcy), triglyceride (TG), total cholesterol (TC), low density lipoprotein (LDL), high density lipoprotein (HDL), insulin resistance (HOMA-IR), malondialdehyde (MDA) and anxiety/depression were all the influencing factors of GDM combined with gestational hypertension (OR=2.776, 2.149, 5.008, 4.586, 4.208, 4.047, 3.473, 0.503, 3.688, 3.317, 2.305, P < 0.05). The above influencing factors were used as predictive indicators to construct a nomogram model for predicting the risk of GDM complicated with pregnancy-induced hypertension. The consistency index of the modeling group and the validation group was 0.830 and 0.827, respectively, and the calibration curve and ideal curve fit were good. The receiver operating characteristic curve evaluation model showed that the area under the curve of the modeling group and the validation group were 0.859 and 0.850, respectively; the decision curve showed that the modeling group and the validation group could obtain higher net benefit values when the risk threshold probability was 0.02 to 0.89 and 0.02 to 0.85, respectively.
Conclusion Prepregnancy overweight/obesity, excessive body mass gain during pregnancy, abnormalities in three indicators of oral glucose tolerance test, Hcy, TG, TC, LDL, HDL, HOMA-IR, MDA and anxiety/depression are all influencing factors of GDM combined with gestational hypertension. The column-line model constructed based on this analysis has good predictive efficacy and can guide clinical screening of high-risk groups.