To evaluate the diagnostic effectiveness and predictive value of fluorescence in situ hybridization (FISH) in the surveillance of non-muscle invasive bladder cancer (NMIBC). PubMed/Medline, EMBASE, Web of Science, Ovid, Web of Knowledge, and Cochrane Library will be searched for studies related to the topic. The identification, inclusion, and exclusion flowcharts will be conducted according to preferred reporting items for systematic reviews and meta-analysis guidelines. The identified reports will be critically appraised according to the Newcastle-Ottawa scale, quality assessment of diagnostic accuracy studies-2 and standards for reporting of diagnostic accuracy 2015. Forest plots will be generated to display hazard ratios, sensitivities, and specificities. Pooled estimates with their 95% confidence intervals will be calculated using the bivariate model, the hierarchical summary receiver operating characteristic model and a fixed- or random-effects model. This study will provide evidence and data to form a comprehensive understanding of the value of FISH in the surveillance of NMIBC. The diagnostic efficacy of FISH will be affected by post-therapy factors. However, FISH still could facilitate the surveillance of NMIBC owing to its non-invasive feature. This study will improve the clinical decision-making and enlighten the future research of NMIBC.