Construction And Analysis Of A Nomogram Model For Predicting The Risk Of Central Precocious Puberty In Obese Girls

Objective:In this study,by analyzing and screening the risk factors of central precocious puberty in obese girls,a nomogram model was constructed to predict the risk of central precocious puberty in obese girls,and the predictive ability of the model was tested.In order to provide a new assessment tool for early clinical identification of the risk of central precocious puberty in obese girls.Methods:A total of 154 obese girls with precocious puberty who were admitted to the Department of Endocrinology,Genetics and Metabolism of Jiangxi Provincial Children’s Hospital from March 2020 to December 2022 and underwent Gn RH stimulation test were selected,and 765 non-obese girls with precocious puberty were selected during the same period.Basic data,growth and development indicators,relevant laboratory indicators and imaging results were collected,and patients were grouped according to the diagnostic criteria for central precocious puberty.81 obese girls were included in the CPP group and 73 were included in the non-CPP group.Among the non-obese girls,535 were included in the CPP group and 230 were included in the non-CPP group.According to the data types of the study objects,appropriate statistical methods were selected to conduct univariate analysis and multivariate Logistic regression analysis to determine the progression of obese and non-obese girls as predictors of central precocious puberty,and the Logistic regression equation was used to establish the prediction model.The predictive ability of the model was evaluated by the area under the receiver operating characteristic curve.The risk prediction model of central precocious puberty in obese girls was visualized with a nomogram model,and the efficacy of the nomogram model for predicting central precocious puberty in obese girls was evaluated and verified by cross validation,calibration curve and clinical decision analysis curve.Results:Univariate analysis showed that there were statistically significant differences between CPP and non-CPP patients in obese and non-obese girls in age of visit,E₂,LH base value,FSH base value,bone age,uterine volume,average follicle volume and mature follicle.The difference in disease duration and breast stage was statistically significant only between CPP and non-CPP patients in non-obese girls.The results of multivariate Logistic regression analysis showed that,The independent risk factors for progression to CPP in obese girls were LH base value（OR=7.47,95%CI 2.01-27.74）,FSH base value（OR=1.74,95%CI 1.15-2.63）and uterine volume（OR=3.15,95%CI 1.84-5.38）.Independent risk factors for progression to CPP in non-obese girls were LH base value（OR=6.92,95%CI 4.20-11.39）,bone age（OR=1.40,95%CI 1.24-1.58）,and uterine volume（OR=1.38,95%CI 1.16-1.65）.According to the independent variables and their coefficients and constants selected by appellate stepwise regression analysis,the Logistic regression prediction model was established.The regression equation of the prediction model of central precocious puberty in obese girls was P=1/1+EXP（-4.38+2.01*LH base value+0.55*FSH base value+1.15*uterine volume）.The regression equation for predicting the risk of central precocious puberty in non-obese girls was P=1/1+EXP（-3.23+1.93*LH base value+0.34*bone age+0.33*uterine volume）.The AUC of the risk prediction model for obese girls was 0.90（95%CI 0.84-0.95）,which was better than that for non-obese girls（95%CI 0.78-0.84）,showing good predictive efficacy.A nomogram model was used to visualized the risk prediction model for central precocious puberty in obese girls.The average AUC of the model after 10-fold cross-validation was 0.88,and the Hosmer-Lemeshow test result showedχ²=5.04（P=0.75）,indicating that this model had good differentiation and calibration.The clinical decision analysis curve suggests that this model has a higher net benefit in clinical use.Conclusions:（1）LH basal value,FSH basal value and uterine volume are independent risk factors for central precocious puberty in obese girls.（2）The prediction efficiency of the nomogram for the risk prediction of central precocious puberty in obese girls constructed in this study is good and at medium level,which can be used to identify high-risk groups of central precocious puberty in obese girls and provide individualized prevention programs.