医学
四分位间距
降钙素原
置信区间
肺炎
一致性
内科学
肺超声
算法
随机对照试验
重症监护室
胃肠病学
败血症
计算机科学
作者
Carmina Guitart,Javier Rodríguez‐Fanjul,Sandra Martínez Pérez,Josep L. Carrasco,Emilio J. Inarejos Clemente,Francisco José Cambra,Mònica Balaguer,Iolanda Jordán
摘要
Lung ultrasound (LUS) and procalcitonin (PCT) are independently used to improve accuracy when diagnosing lung infections. The aim of the study was to evaluate the accuracy of a new algorithm combining LUS and PCT for the diagnosis of bacterial pneumonia.Randomized, blinded, comparative effectiveness clinical trial. Children <18 years old with suspected pneumonia admitted to pediatric intensive care unit were included, and randomized into experimental group (EG) or control group (CG) if LUS or chest X-Ray (CXR) were done as the first pulmonary image, respectively. PCT was determined. In patients with bacterial pneumonia, sensitivity, specificity, and predictive values of LUS, CXR, and of both combined with PCT were analyzed and compared. Concordance between the final diagnosis and the diagnosis concluded through the imaging test was assessed.A total of 194 children, with a median age of 134 (interquartile range [IQR]: 39-554) days, were enrolled, 96 randomized into the EG and 98 into the CG. Bacterial pneumonia was diagnosed in 97 patients. Sensitivity and specificity for bacterial pneumonia diagnosis were 78% (95% confidence interval [CI]: 70-85) and 98% (95% CI: 93-99) for LUS, 85% (95% CI: 78-90) and 53% (95% CI: 43-62) for CXR, 90% (95% CI: 83-94) and 85% (95% CI: 76-91) when combining LUS and PCT, and 95% (95% CI: 90-98) and 41% (95% CI: 31-52) when combining CXR and PCT. The positive predictive value for LUS and PCT was 88% (95% C:I 79%-93%) versus 68% (95% CI: 60-75) for CXR and PCT. The concordance between the final diagnosis and LUS had a kappa value of 0.69 (95% CI: 0.62-0.75) versus 0.34 (95% CI: 0.21-0.45) for CXR, (p < 0.001).The combination of LUS and PCT presented a better accuracy for bacterial pneumonia diagnosis than combining CXR and PCT. Therefore, its implementation could be a reliable tool for pneumonia diagnosis in critically ill children.
科研通智能强力驱动
Strongly Powered by AbleSci AI