A novel congenital diaphragmatic hernia prediction model for Chinese subjects: A multicenter cohort investigation

PURPOSE OF THE STUDY: Brindle et al. (2014) and the Congenital Diaphragmatic Hernia Study Group constructed a simplified clinical prediction rule (Brindle score) to stratify infants with congenital diaphragmatic hernia based on disease severity. We aimed to develop a predictive model applicable to Chinese patients with congenital diaphragmatic hernia and externally validate whether the Brindle score is applicable to the Chinese population. PATIENTS AND THE METHODS: Multiple imputations supplemented the missing data. A least absolute shrinkage and selection operator regression was used to screen the factors influencing adverse outcomes. Internal validation was performed by bootstrap resampling. The C-index, area under the receiver operating characteristic curve, and the Hosmer-Lemeshow test evaluated the predictive power. RESULTS: A nomogram named “CCDH score” (Chinese Congenital Diaphragmatic Hernia score), including pulmonary hypertension, low 5-min Apgar score (<7), chromosomal anomaly, major cardiac anomalies (MCAs), observed-to-expected lung-to-head ratio, and the percentage of liver herniation, was constructed. The CCDH score revealed good calibration and discriminative abilities, with a C-index of 0.941. In the training and external validation cohorts, the area under the receiver operating characteristic curve of the Brindle score were 0.820 and 0.881, respectively. The Brindle score has fair predictive power in the Chinese population, but the newly established CCDH score seems more suitable for Chinese patients with congenital diaphragmatic hernia. CONCLUSION: The CCDH score is the first predictive model constructed based on the characteristics of the Chinese population and can accurately predict the survival outcomes of patients with congenital diaphragmatic hernia.