Background: Protocols of systematic reviews allow for planning and documentation of review methods, and thus improve the transparency of reviews process. However, pre-registered a protocol is not enough, the author also need to follow it. PROSPERO is an open-access online facility for the registration of non-Cochrane systematic reviews. The purpose of our research is to determined what changed were made between non-Cochrane reviews and their protocols in PROSPERO and how likely these changes impacted the quality of systematic review. Method: In this retrospective comparative study we electronically searched for protocols and their corresponding systematic reviews in the PROSPERO platform that were “completed and published” from January to December, 2018. Two reviewers independently identified and classified changes between the protocols and systematic reviews then evaluated the impact (improve/reduce/unclear)of these change on the reporting/methodology quality of reviews. Frequency (n), percentage (%) were used to analyze the number of changes categorically in each review and the distribution of different impact caused by these changes. Results: We identified 39 pre-registered protocols and their reviews, all of which exhibited alterations. All changes to only one review are considered to improve the reporting/methodology quality, and remaining 97% of reviews (n=38) contain changes that are categorically considered to reduce the methodology/reporting quality or that have an unclear impact on reviews. Conclusions: Differences between the non-Cochrane reviews and their protocols recorded in PROSPERO are widespread, and there have been many changes having an unclear impact on the quality of reviews. Guiding the author to report and explain the differences between protocol and reviews or even requiring authors to so at the level of journal are two fundamental solutions to further improve the transparency of the non-Cochrane reviews.