Email Alert | RSS

Chinese Journal of Antituberculosis ›› 2022, Vol. 44 ›› Issue (8): 844-848.doi: 10.19982/j.issn.1000-6621.20220065

• Original Articles • Previous Articles     Next Articles

Evaluation of the potential of cross-priming amplification for the diagnosis of Mycobacterium tuberculosis infection

Jiang Shuping1, Liu Changwei2(), Li Bin1, Xu Xiaoqin1, Luo Wenji1, Yu Bolin1, Zhang Yue1   

  1. 1Laboratory of Shenzhen Longhua District Chronic Disease Prevention and Control Center, Shenzhen 518110, China
    2Office of Shenzhen Longhua District Chronic Disease Prevention and Control Center, Shenzhen 518110, China
  • Received:2022-03-07 Online:2022-08-10 Published:2022-08-03
  • Contact: Liu Changwei E-mail:JSPapple0755@163.com
  • Supported by:
    Special Fund for Talents of Science and Technology Innovation Bureau of Longhua District, Shenzhen(2020051)

Abstract:

Objective: To evaluate the potential of cross-priming amplification (CPA) in the diagnosis of patients newly diagnosed with suspected pulmonary tuberculosis. Methods: Of 2408 new cases of suspected pulmonary tuberculosis presenting at Longhua District Chronic Disease Control Center in Shenzhen City from January, 2020 to May, 2021 were enrolled in the study. Sputum samples were collected and sputum smears (Ziehl-Neelsen acid-fast stain (AFS) method), BACTEC MGIT 960 (MGIT 960) liquid culture and CPA analysis were performed to detect Mycobacterium tuberculosis (MTB), MGIT 960 liquid culture being used as the gold standard to evaluate the diagnostic efficacy of CPA. Results: The MTB detection rate by AFS, MGIT 960 liquid culture and CPA was 8.89% (214/2408), 27.99% (674/2408) and 15.49% (373/2408), respectively. The difference in the detection rates of the three methods was statistically significant (χ2=314.619, P=0.000), with that of MGIT 960 liquid culture being significantly higher than both CPA and AFS, and that of CPA being higher than that of AFS (χ2=110.572, 292.158, 49.046, P=0.000). Using MGIT 960 liquid culture as the gold standard, the sensitivity, specificity, positive predictive value, negative predictive value and coincidence rate of CPA in detecting MTB were 52.37% (353/674), 98.85% (1714/1734), 94.64% (353/373), 84.23% (1714/2035) and 85.84% (2067/2408), respectively, while that of the AFS method was 31.45% (212/674), 99.88% (1732/1734), 99.07% (212/214), 78.94% (1732/2194) and 80.73% (1944/2408), respectively. The sensitivity of CPA (52.37%; 95%CI: 47.21%-59.64%) was significantly higher than that of AFS (31.45%; 95%CI: 26.87%-35.72%;χ2=108.290, P=0.000). Results of CPA were in good agreement with MGIT 960 liquid culture (Kappa=0.592), but AFS results showed poor agreement (Kappa=0.395). The smear-positive/culture-negative rate was 0.08% (2/2408), and the culture contamination rate was 5.40% (130/2408). All CPA detection results were valid. Conclusion: The positive detection rate and sensitivity of CPA for diagnosing cases of suspected MTB infection was significantly higher than that of AFS, and showed better consistency. The CPA technique is simple, rapid, economical and free of contamination, making it of promising value in the early diagnosis of MTB infection.

Key words: Tuberculosis,pulmonary, Nucleic acid amplification techniques, Clinical laboratory techniques, Diagnosis, Comparative study

CLC Number: