Respiratory viruses are recognized as serious causes of morbidity and mortality in lower respiratory tract infections in patients. Isothermal amplification assays are increasingly used in their detection because of their rapidity, simplicity and cost-effectiveness, when compared to traditional molecular diagnostic methods. However, systematic assessment of these assays in the clinical settings is rarely reported. MEDLINE (Pubmed) searches were done analysing subject headings and words in the abstract related to isothermal amplification assay and virus. Selected loop-mediated isothermal amplification assays (LAMP) for respiratory syncytial virus (RSV), human metapneumovirus (HMPV) and adenovirus (ADV) as well as a reverse transcription genome exponential amplification reaction assay (GEAR) for human rhinovirus (HRV) were clinically evaluated in a head to head comparison against a two-tube multiplex reverse transcription-PCR (RT-PCR) assay (two-tube assay) using 634 respiratory specimens from children with pneumonia from different regions in China. Discrepant results between isothermal amplification assays and the two-tube assay were resolved by sequencing. A comparison of sensitivities of each selected isothermal amplification assay among province, gender, and age groups was also analyzed. A total of 634 respiratory specimens selected from Hebei Province children's hospital and Hunan Provincial Center for Disease Control and Prevention were tested. The overall detection rate (number of positive specimens/total specimens) for viruses tested by Reverse transcription (RT)-LAMP/LAMP/RT-GEAR was 35.9% while the detection rate was 46.2% by the two-tube assay. The sensitivity of each isothermal amplification assay was 88.4%, 74.3%, 100% and 73.6% for RSV, HMPV, ADV and HRV, respectively. No false positives were found in isothermal amplification assays. All the discrepant negative results by isothermal amplification assays were confirmed false negatives by sequencing. The LAMP assay for ADV showed significant consistency with the two-tube assay. A higher sensitivity of RSV detection was found in Hunan Province than in Hebei Province (P = 0.01). Among different age groups, a higher sensitivity of RSV detection was also found in children older than 1 year, when compared to children less than 1 year (P = 0.01). The clinical performance of the selected isothermal amplification assays for different viruses varies. Multiple-center assessment is critical to evaluate isothermal amplification assays, especially for RNA viruses, for their broad use in clinical hospital. The selected LAMP assay for the detection of ADV is reliable and has great potential for use in clinics; however, the sensitivities of the LAMP/GEAR assays for the detection of RSV, HMPV and HRV need to be further improved to meet clinical requirements, although a statistical difference in sensitivity was only found for the selected LAMP assay for RSV.