OBJECTIVE: To determine the diagnostic accuracy of lateral neck radiography and its value in related radiographic findings in the detection of foreign bodies using the likelihood ratio. PATIENTS AND METHODS: This retrospective study comprised 114 patients who had neck radiography followed by rigid esophagoscopy. Interpretations of the radiographs were correlated with esophagoscopic findings and clinical follow-up for at least 1 month. Data were mainly assessed by likelihood ratio and detection rate. RESULTS: Plain radiography had a positive likelihood ratio of 2.4 (95% confidence interval [CI] 1.8-3.1) and a negative likelihood ratio of 0.6 (95% CI 0.40-0.7). It had a sensitivity of 57.4% (95% CI 45.6-69.1) and a specificity of 76.1% (95% CI 63.8-88.4). Positive likelihood ratios for each radiographic finding (foreign body shadow, air, and soft tissue swelling) were 3.4 (95% CI 2.6-4.4), 1.5 (95% CI 0.9-2.6), and 4.1 (95% CI 1.9-8.7), respectively. Plain radiography had a high detection rate (72.5%) when foreign bodies were lodged in the hypopharynx and cervical esophagus. On the contrary, the detection rate was poor (0%) in the oropharynx. CONCLUSIONS: A lateral neck radiograph is better taken after a negative finding in a thorough oral examination and fibre-optic laryngoscopy. Judging from the possible site of impacted foreign body and radiographic findings, we can better predict the presence of foreign bodies.
OBJECTIVE: To determine the diagnostic accuracy of lateral neck radiography and its value in related radiographic findings in the detection of foreign bodies using the likelihood ratio. PATIENTS AND METHODS: This retrospective study comprised 114 patients who had neck radiography followed by rigid esophagoscopy. Interpretations of the radiographs were correlated with esophagoscopic findings and clinical follow-up for at least 1 month. Data were mainly assessed by likelihood ratio and detection rate. RESULTS: Plain radiography had a positive likelihood ratio of 2.4 (95% confidence interval [CI] 1.8-3.1) and a negative likelihood ratio of 0.6 (95% CI 0.40-0.7). It had a sensitivity of 57.4% (95% CI 45.6-69.1) and a specificity of 76.1% (95% CI 63.8-88.4). Positive likelihood ratios for each radiographic finding (foreign body shadow, air, and soft tissue swelling) were 3.4 (95% CI 2.6-4.4), 1.5 (95% CI 0.9-2.6), and 4.1 (95% CI 1.9-8.7), respectively. Plain radiography had a high detection rate (72.5%) when foreign bodies were lodged in the hypopharynx and cervical esophagus. On the contrary, the detection rate was poor (0%) in the oropharynx. CONCLUSIONS: A lateral neck radiograph is better taken after a negative finding in a thorough oral examination and fibre-optic laryngoscopy. Judging from the possible site of impacted foreign body and radiographic findings, we can better predict the presence of foreign bodies.