TY - JOUR
T1 - Intra- and Interobserver Agreement Among Bronchial Endosonographers for the Description of Intrathoracic Lymph Nodes
AU - Garcia-Olivé, Ignasi
AU - Radua, Joaquim
AU - Serra, Pere
AU - Andreo, Felipe
AU - Sanz-Santos, Jose
AU - Monsó, Eduard
AU - Rosell, Antoni
AU - Cases-Viedma, Enrique
AU - Fernández-Villar, Alberto
AU - Núñez-Delgado, Manuel
AU - García-Luján, Ricardo
AU - Morera, Josep
AU - Ruiz-Manzano, Juan
PY - 2012/7/1
Y1 - 2012/7/1
N2 - Several sonographic features observed by endobronchial ultrasonography have been suggested to be useful to predict malignancy in mediastinal lymph nodes. To evaluate agreement to describe sonographic features, 28 video images were evaluated twice by eight expert bronchoscopists. The observers reviewed each case for the presence of coagulation necrosis sign (CNS), central hilar structure (CHS), heterogeneity, distinct margin, round shape, size >1 cm and malignancy. Intraobserver agreement was almost perfect for size (κ = 0.826), substantial for CNS (κ = 0.721) and shape (κ = 0.615), and moderate for CHS (κ = 0.565), heterogeneity (κ = 0.441) and margin (κ = 0.407). Interobserver agreement was substantial for size (κ = 0.641), moderate for shape (κ = 0.445), and fair for CNS (κ = 0.340) and margin (κ = 0.274). In conclusion, inter- and intraobserver agreement of the endosonographic features for mediastinal or hilar lymph nodes is good for shape or size but not good enough for the other ultrasonographic features. © 2012 World Federation for Ultrasound in Medicine & Biology.
AB - Several sonographic features observed by endobronchial ultrasonography have been suggested to be useful to predict malignancy in mediastinal lymph nodes. To evaluate agreement to describe sonographic features, 28 video images were evaluated twice by eight expert bronchoscopists. The observers reviewed each case for the presence of coagulation necrosis sign (CNS), central hilar structure (CHS), heterogeneity, distinct margin, round shape, size >1 cm and malignancy. Intraobserver agreement was almost perfect for size (κ = 0.826), substantial for CNS (κ = 0.721) and shape (κ = 0.615), and moderate for CHS (κ = 0.565), heterogeneity (κ = 0.441) and margin (κ = 0.407). Interobserver agreement was substantial for size (κ = 0.641), moderate for shape (κ = 0.445), and fair for CNS (κ = 0.340) and margin (κ = 0.274). In conclusion, inter- and intraobserver agreement of the endosonographic features for mediastinal or hilar lymph nodes is good for shape or size but not good enough for the other ultrasonographic features. © 2012 World Federation for Ultrasound in Medicine & Biology.
KW - Agreement
KW - EBUS
KW - Interobserver agreement
KW - Intraobserver agreement
KW - Observer dependency
KW - Sonographic features
U2 - 10.1016/j.ultrasmedbio.2012.03.012
DO - 10.1016/j.ultrasmedbio.2012.03.012
M3 - Article
SN - 0301-5629
VL - 38
SP - 1163
EP - 1168
JO - Ultrasound in Medicine and Biology
JF - Ultrasound in Medicine and Biology
IS - 7
ER -