Guocan Yu1, Fangming Zhong2, Wuchen Zhao3, Bo Ye4, Kan Xu5, Gang Chen6. 1. Zhejiang Tuberculosis Diagnosis and Treatment Center, Affiliated Hangzhou Chest Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China. Electronic address: dabaitwo@163.com. 2. Zhejiang Tuberculosis Diagnosis and Treatment Center, Affiliated Hangzhou Chest Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China. Electronic address: zhongfangm@126.com. 3. Zhejiang Tuberculosis Diagnosis and Treatment Center, Affiliated Hangzhou Chest Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China. Electronic address: 773147656@qq.com. 4. Zhejiang Tuberculosis Diagnosis and Treatment Center, Affiliated Hangzhou Chest Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China. Electronic address: yeboboye@126.com. 5. Zhejiang Tuberculosis Diagnosis and Treatment Center, Affiliated Hangzhou Chest Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China. Electronic address: xukanmed@163.com. 6. Zhejiang Tuberculosis Diagnosis and Treatment Center, Affiliated Hangzhou Chest Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China. Electronic address: chengangc@126.com.
Abstract
BACKGROUND: The diagnosis of constrictive tuberculous pericarditis (TBP) remains challenging. This study aimed to evaluate 5 tests (acid-fast bacilli [AFB] smear, Mycobacterium tuberculosis [MTB] culture, Xpert MTB/RIF assay, CapitalBio Mycobacterium real-time PCR detection assay [CapitalBio assay], and pathology) for constrictive TBP using pericardial tissue. METHODS: We reviewed the case histories of patients with suspected constrictive TBP. We analyzed the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the curve (AUC) of these assays. RESULTS: A total of 69 patients were included. The sensitivity, specificity, PPV, NPV, and AUC of AFB smear were 7.3%, 100.0%, 100.0%, 21.5%, and 0.54, respectively; those of culture were 23.6%, 100.0%, 100.0%, 25.0%, and 0.62, respectively; those of Xpert MTB/RIF were 52.7%, 100.0%, 100.0%, 35.0%, and 0.76, respectively; those of CapitalBio assay were 50.9%, 100.0%, 100.0%, 34.2%, and 0.75, respectively; and those of pathology were 92.7%, 92.9%, 98.1%, 76.5%, and 0.93, respectively. CONCLUSIONS: The validity of AFB smear and MTB culture remains low. Nucleic acid amplification tests can provide diagnostic efficacy for TBP but only moderately. The CapitalBio assay and Xpert MTB/RIF were considered similar for diagnosing TBP. Pathology showed the best diagnostic accuracy among the 5 tests.
BACKGROUND: The diagnosis of constrictive tuberculous pericarditis (TBP) remains challenging. This study aimed to evaluate 5 tests (acid-fast bacilli [AFB] smear, Mycobacterium tuberculosis [MTB] culture, Xpert MTB/RIF assay, CapitalBio Mycobacterium real-time PCR detection assay [CapitalBio assay], and pathology) for constrictive TBP using pericardial tissue. METHODS: We reviewed the case histories of patients with suspected constrictive TBP. We analyzed the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the curve (AUC) of these assays. RESULTS: A total of 69 patients were included. The sensitivity, specificity, PPV, NPV, and AUC of AFB smear were 7.3%, 100.0%, 100.0%, 21.5%, and 0.54, respectively; those of culture were 23.6%, 100.0%, 100.0%, 25.0%, and 0.62, respectively; those of Xpert MTB/RIF were 52.7%, 100.0%, 100.0%, 35.0%, and 0.76, respectively; those of CapitalBio assay were 50.9%, 100.0%, 100.0%, 34.2%, and 0.75, respectively; and those of pathology were 92.7%, 92.9%, 98.1%, 76.5%, and 0.93, respectively. CONCLUSIONS: The validity of AFB smear and MTB culture remains low. Nucleic acid amplification tests can provide diagnostic efficacy for TBP but only moderately. The CapitalBio assay and Xpert MTB/RIF were considered similar for diagnosing TBP. Pathology showed the best diagnostic accuracy among the 5 tests.