The correlation between lung ultrasound score and neonatal critical illness score and its clinical predictive value / 中华超声影像学杂志
Chinese Journal of Ultrasonography
; (12): 748-752, 2019.
Article
de Zh
| WPRIM
| ID: wpr-791291
Bibliothèque responsable:
WPRO
ABSTRACT
Objective To explore whether the lung ultrasound( LUS) score can be used to assess and predict the criticality of neonates with pulmonary disease at an early stage . Methods T he new borns born in the obstetrics department of Affiliated Hospital of Jining M edical University from April to October 2018 were transferred to the neonatal intensive care unit due to respiratory distress . T he children underwent LUS examination and scoring at 2 hours after birth . T he correlation analysis were performed between LUS score and neonatal critical illness score ( NCIS ) ,NCIS +single index ,respectively . And the ROC curve was used to analyze the value of LUS score in predicting neonatal criticality . Results ①T he LUS score of non‐critical neonates was significantly lower than that of critically ill newborns , the difference was statistically significant ( P =0 .005) ; LUS score was an independent risk factor for critical neonates ( OR=1 .71 ,95%CI :1 .059-2 .765 , P = 0 .028 ) . ② T he correlation coefficient between LUS score and NCIS was -0 .48 ( P =0 .002) . T he correlation coefficient between the LUS score and the NCIS + single index was -0 .44 ( P=0 .005) . ③T he area under the ROC curve of LUS score predicting neonatal criticality was 0 .88 ( 95%CI :0 .725-0 .965 , P <0 .000 1) ,the optimal diagnostic threshold was 6 points with sensitivity of 80% and specificity of 100% . Conclusions The LUS score at a postnatal age of 2 hours after birth can early assess and predict the criticality of neonates with pulmonary disease . And the LUS score greater than 6 has the highest diagnostic value .
Texte intégral:
1
Base de données:
WPRIM
Type d'étude:
Prognostic_studies
/
Risk_factors_studies
Langue:
Zh
Journal:
Chinese Journal of Ultrasonography
Année:
2019
Type de document:
Article