BACKGROUND : A novel method, termed loop-mediated isothermal amplification (LAMP), was developed by Notomi et al. (2000). Individually published results have been reported that this technology has been successfully applied to the detection of a variety of pathogens. However, the overall diagnostic accuracy of LAMP for Mycoplasma pneumoniae (MP) remains unclear. A meta-analysis was therefore performed to review the accuracy of LAMP for Mycoplasma pneumoniae. METHODS : Cochrane Library and PubMed were systematically searched and checked for studies using LAMP for detecting mycoplasma pneumoniae. We used PCR as a reference standard to evaluate the quality of the studies eligible for inclusion in the meta-analysis. Then, the data from the studies were extracted by two independent assessors. Meta-DiSc 1.4 software was utilized to test the heterogeneity of sensitivity (SEN), specificity (SP), positive likelihood ratio (PLR), negative likelihood ratio (NLR), and diagnosis odds ratio (DOR). The pooled analysis results were plotted, and the summary receiver operating characteristic (SROC) curve was plotted by calculating the area under the curve (AUC). Generated pooled summary estimates (95% CIs) were calculated for the overall accuracy, and a bivariate meta-regression model was used for the meta-analysis. RESULTS : Seven studies with nine fourfold tables were included in this meta-analysis. The pooled SEN and SPE for diagnosing Mycoplasma pneumoniae were 0.90 (95% CI: 0.87-0.93) and 0.98 (95% CI: 0.96-0.99), respectively. The PLR was 31.25 (95% CI: 14.83-65.87), NLR 0.10 (95% CI: 0.05-0.22), DOR 399.32 (95% CI: 172.01-927.00), and AUC 0.9892. CONCLUSIONS : In conclusion, compared with PCR, LAMP is a valuable alternative method for Mycoplasma pneumoniae diagnosis in clinic with high sensitivity and specificity. However, more evidence is required to confirm that LAMP can fully replace other methods in the clinical diagnosis of MP.