1. Endobronchial Ultrasound for Nodal Staging of Patients with Non-Small-Cell Lung Cancer with Radiologically Normal Mediastinum. A Meta-Analysis
- Author
-
Hazem E El-Osta, Philip A. Rascoe, Pushan P Jani, Syed H Jafri, and Ali Mansour
- Subjects
Pulmonary and Respiratory Medicine ,medicine.medical_specialty ,Lung Neoplasms ,Nodal staging ,030204 cardiovascular system & hematology ,Endosonography ,03 medical and health sciences ,0302 clinical medicine ,Carcinoma, Non-Small-Cell Lung ,Bronchoscopy ,medicine ,Humans ,Radiation treatment planning ,Lung cancer ,Neoplasm Staging ,Lung ,business.industry ,Mediastinum ,medicine.disease ,medicine.anatomical_structure ,030228 respiratory system ,Mediastinal lymph node ,Meta-analysis ,Lymphatic Metastasis ,Radiography, Thoracic ,Non small cell ,Radiology ,Lymph Nodes ,business - Abstract
An accurate assessment of the mediastinal lymph node status is essential in the staging and treatment planning of potentially resectable non-small-cell lung cancer.We performed this meta-analysis to evaluate the role of endobronchial ultrasound-guided transbronchial needle aspiration in detecting occult mediastinal disease in non-small-cell lung cancer with no radiologic mediastinal involvement.The PubMed, Embase, and Cochrane libraries were searched for studies describing the role of endobronchial ultrasound-guided transbronchial needle aspiration in patients with lung cancer with radiologically negative mediastinum. The individual and pooled sensitivity, prevalence, negative predictive value, and diagnostic odds ratio were calculated using the random effects model. Meta-regression analysis, heterogeneity, and publication bias were also assessed.A total of 13 studies that met the inclusion criteria were included in the meta-analysis. The pooled effect sizes of the different diagnostic parameters were estimated as follows: prevalence, 12.8% (95% confidence interval, 10.4-15.7%); sensitivity, 49.5% (95% confidence interval, 36.4-62.6%); negative predictive value, 93.0% (95% confidence interval, 90.3-95.0%); and log diagnostic odds ratio, 5.069 (95% confidence interval, 4.212-5.925). Significant heterogeneity was noticeable for the sensitivity, disease prevalence, and negative predictive value, but not observed for log diagnostic odds ratio. Publication bias was detected for sensitivity, negative predictive value, and log diagnostic odds ratio but not for prevalence. Bivariate meta-regression analysis showed no significant association between the pooled calculated parameters and the type of anesthesia, imaging used to define negative mediastinum, rapid on-site test usage, and presence of bias by Quality Assessment of Diagnostic Accuracy Studies (QUADAS)-2 tool. Interestingly, we observed a greater sensitivity, negative predictive value, and log diagnostic odds ratio for studies published before 2010 and for prospective multicenter studies.Among patients with non-small-cell lung cancer with a radiologically normal mediastinum, the prevalence of mediastinal disease is 12.8% and the sensitivity of endobronchial ultrasound-guided transbronchial needle aspiration is 49.5%. Despite the low sensitivity, the resulting negative predictive value of 93.0% for endobronchial ultrasound-guided transbronchial needle aspiration suggests that mediastinal metastasis is uncommon in such patients.
- Published
- 2018