In S Moon1, Dong W Kim, Hye J Baek. 1. Department of Radiology , Busan Paik Hospital, Inje University School of Medicine, Busan, South Korea.
Abstract
OBJECTIVES/HYPOTHESIS: No previous ultrasound (US) studies have investigated the diagnostic accuracy of specific US-based diagnostic protocol for cervical lymph node (LN) lesions in a tuberculosis-endemic area. This study aimed to assess the accuracy of US-based diagnosis of cervical LNs in a tuberculosis-endemic area. STUDY DESIGN: This was a retrospective cohort study. METHODS: We enrolled 495 consecutive patients who underwent US examination of the LNs in the neck. A single radiologist carried out the US examination, and established a US-based diagnosis at the time of the examination in each of the patients. We calculated the diagnostic accuracy of US for LNs in the neck by using the final results as a reference standard. RESULTS: Of 476 study cases, the number of tuberculous, nontuberculous benign, malignant, and indeterminate US category cases was 74, 224, 152, and 26, respectively. Confirmed diagnoses in these cases were as follow: tuberculous LN (n = 69), nontuberculous benign LN (n = 249), and malignant LN (n = 158). The sensitivity, specificity, positive and negative predictive values, and accuracy of tuberculous, malignant, and nontuberculous benign US categories were 90.9%, 96.4%, 81.1%, 98.4%, and 95.6%; 92.8%, 96.6%, 93.4%, 96.3%, and 95.3%; and 92.2%, 95.0%, 95.1%, 92.0%, and 93.6%, respectively. All US categories showed a high sensitivity and specificity. CONCLUSIONS: The present US-based diagnostic protocol for cervical LNs in a tuberculosis-endemic area may be useful. LEVEL OF EVIDENCE: 4.
OBJECTIVES/HYPOTHESIS: No previous ultrasound (US) studies have investigated the diagnostic accuracy of specific US-based diagnostic protocol for cervical lymph node (LN) lesions in a tuberculosis-endemic area. This study aimed to assess the accuracy of US-based diagnosis of cervical LNs in a tuberculosis-endemic area. STUDY DESIGN: This was a retrospective cohort study. METHODS: We enrolled 495 consecutive patients who underwent US examination of the LNs in the neck. A single radiologist carried out the US examination, and established a US-based diagnosis at the time of the examination in each of the patients. We calculated the diagnostic accuracy of US for LNs in the neck by using the final results as a reference standard. RESULTS: Of 476 study cases, the number of tuberculous, nontuberculous benign, malignant, and indeterminate US category cases was 74, 224, 152, and 26, respectively. Confirmed diagnoses in these cases were as follow: tuberculous LN (n = 69), nontuberculous benign LN (n = 249), and malignant LN (n = 158). The sensitivity, specificity, positive and negative predictive values, and accuracy of tuberculous, malignant, and nontuberculous benign US categories were 90.9%, 96.4%, 81.1%, 98.4%, and 95.6%; 92.8%, 96.6%, 93.4%, 96.3%, and 95.3%; and 92.2%, 95.0%, 95.1%, 92.0%, and 93.6%, respectively. All US categories showed a high sensitivity and specificity. CONCLUSIONS: The present US-based diagnostic protocol for cervical LNs in a tuberculosis-endemic area may be useful. LEVEL OF EVIDENCE: 4.