RESUMEN
BACKGROUND: Mycoplasma pneumoniae (M pneumoniae) is a common human etiology of respiratory infections. Nuclear acid sequence-based amplification (NASBA) shows good value for the detection of M pneumoniae that surpasses PCR. However, the optimal detection technology still remains to be identified. The purpose of this meta-analysis was to systematically evaluate the overall accuracy of NASBA for diagnosing M pneumoniae infections. METHODS: The databases PubMed, Cochrane Library, Google Scholar, CNKI, Wang Fang, and Baidu Scholar were comprehensively searched from their initiation date to December 2017 for NASBA in the diagnosis of M pneumoniae infection. Meta-DiSc 1.4 statistical software was used to evaluate the sensitivity (SEN), specificity (SPE), negative likelihood ratio (-LR), positive likelihood ratio (+LR), diagnostic odds ratio (DOR), and summary receiver operating characteristic (SROC). RevMan 5.2 statistical software was used for quality evaluation of the included articles. Publication bias was evaluated by funnel plot. RESULTS: Six articles with high quality, including 10 studies, were finally included in this meta-analysis. The combined statistics results for the diagnosis of M pneumoniae infection by NASBA were 0.77 (SEN, 95% CI: 0.71 to 0.82); 0.98 (SPE, 95% CI: 0.98 to 0.99); 0.22 (-LR, 95% CI: 0.13 to 0.39); 50.38 (+ LR, 95% CI: 21.85 to 116.17); 292.72 (DOR, 95% CI: 95.02 to 901.75); and 0.9875 (the area under the curve of SROC). CONCLUSION: Nuclear acid sequence-based amplification is a reliable technique to diagnose M pneumoniae infection. However, whether it can replace PCR and serology need to be further studied.