Semi-quantitative assessment optimized the grading of pulmonary aspiration on salivagram in children

• Published in: Annals of nuclear medicine 2021-03, Vol.35 (3), p.321-327
• Main Authors: Shao, Fuqiang, Zhao, Xiaofei, Toyama, Hiroshi, Ichihara, Takashi, Zhuang, Hongming, Zhao, Ruifang, Kung, Boom Ting, Ng, Kwok Sing, Zhang, Ziwei, Wu, Ha
• Format: Article
• Language: English
• Subjects: Child; Child, Preschool; Children; Classification; Female; Field of view; Humans; Imaging; Infant; Male; Medical imaging; Medicine; Medicine & Public Health; Nuclear Medicine; Original Article; Physicians; Pulmonary aspiration; Radiology; Radionuclide Imaging - methods; Respiratory Aspiration - diagnostic imaging; Respiratory tract; Retrospective Studies; ROC Curve; Saliva; Statistical analysis
• ISSN: 0914-7187; 1864-6433; 1864-6433
• DOI: 10.1007/s12149-020-01564-6

Objective Salivagram is one of the imaging modalities to detect pulmonary aspiration in children. This study aims to optimize the classification of pulmonary aspiration detected by salivagram with a semi-quantitative analytical method. Methods This is a retrospective study involving 737 patients (471 males, 266 females; aged 1 month to 8 years; mean age 5.3 months, median age 3.0 months old) with suspected pulmonary aspiration, who had salivagram done between January 2018 and June 2019. Positive cases were divided into 10 groups (Grade 1, R2, L2, R2L2, R3, L3, R3L2, R2L3, R3L3, and 4) according to the scintigraphic findings. Aspiration index was determined as the ratio of the count in the respiratory tract to the total count in the image field of view and compared among different groups using the receiver operating characteristics (ROC) curve analysis. Results A total of 180 cases had positive scintigraphic findings of various grades of aspiration (24.4%, 180/737). There is a high consistency among the two independent nuclear medicine physicians involved in the study, in determining both the disease gradings (κ = 0.919;95% CI: 0.915–0.923) and aspiration index (ICC = 0.994;95% CI: 0.993–0.996). There is no significant difference ( p  > 0.05) in aspiration index among the gradings in “mild” group (grade 1, R2, L2, L2R2), and “moderate” group (grade R3, R3L2, R3L3). After dividing the different grades into “mild”, “moderate” and “severe” groups, the aspiration index of “mild” group is 4.40 ± 2.01, that of “moderate” group is 16.43 ± 8.20, and that of “severe” group is 46.94 ± 14.81. Difference in groups was statistically significant ( p