Abstract Study question Can the artificial removal of the zona pellucida (ZP) at the pronuclear (PN)-stage prevent severe fragmentation in early-cleavage embryos and improve good-quality blastocyst development rate? Summary answer The artificial removal of the ZP at the 2PN-stage could prevent significantly severe fragmentation in early-cleavage embryos and improved the good-quality blastocyst development rate. What is known already Recent analysis using time-lapse culture demonstrated that fragments during early-cleavage occur at the site of the perivitelline threads between the ZP and the embryo membrane (Reprod Biomed Online 2017;35:640-5,646-56). Based on these findings we conducted a pilot study on embryonic development after ZP-removal using abnormally-fertilized oocytes (J Assist Reprod Gene2020;37:1349-54). We showed that the inter-blastomere adhesion was maintained and fragmentation was significantly lower in the ZP-free embryos than in the ZP-intact embryos. Therefore, we attempted to apply this method to patients who have given full consent and had repeatedly failed to conceive due to severe fragmentation in early cleavage-stage embryos. Study design, size, duration This is a single-centre retrospective exploratory study of 34 patients that were scheduled to undergo ART-treatment in our clinic between February 2020 and January 2021. They had been previously treated in our facilities (January 2016 to January 2020) with less than 10% morphologically good-quality blastocysts due to severe fragmentation. None of these patients became pregnant during their initial treatments. Totally, 173 2PN-zygotes were obtained, with 101 allocated to ZP-free group and 72 to ZP-intact group. Participants/materials, setting, methods Normally-fertilized oocytes were placed in 0.125M sucrose-containing HEPES-medium (approximately 5min) and cultured under: 1) ZP-free, in which the ZP was completely removed from the ooplasm by laser-irradiation and a medium-blowing method with a biopsy-pipette, or 2) ZP-intact. ZP-free and ZP-intact zygotes were cultured in sequential medium. Embryos were either freshly transferred on Day-2(ZP-intact group only, according to patients’ wishes), or Day-5/6, or they were cryopreserved on Day-5/6/7 for future embryo transfer cycles for both groups. Main results and the role of chance Even allocation of the normally fertilised oocytes to the two groups was not possible, due to the patients’ request to include more zygotes in the ZP-free group. Therefore, no specific criteria were set other than the rule that all the normally fertilised oocytes of each patient must be allocated into both groups at the time of allocation. As a result, each patient had 2PN zygotes assigned to both ZP-free and ZP-intact groups. Rates of good-quality embryos, blastocyst development, morphologically good-quality blastocyst development and cryopreservation demonstrated significant improvement (all P < 0.01) in the ZP-free group compared to the ZP-intact group. Pregnancy rates in ZP-intact group were 5.9% per patient and 5.3% per embryo transferred, while in ZP-free group 37.5% per patient and 24.3% per embryo transferred. Our results demonstrate that in difficult-to-treat cases where embryos with good morphology cannot be obtained due to excessive fragmentation at the early-cleavage, embryonic development can be significantly improved by artificial removal of ZP at the 2PN-stage. We believe that the ZP-free method has opened a new door in assisted reproductive medicine and can provide the opportunity to obtain blastocysts with good-morphology from an embryo that would probably not have become a blastocyst otherwise. Limitations, reasons for caution A comparative analysis between the two groups was not possible due to the small number of patients, the inclusion of fresh embryos as well as frozen embryos, and the non-randomized data. The imbalance was attributed to the fact that patients wanted us to allocate more oocytes to the ZP-free group. Wider implications of the findings The development of treatment methods based on new ideas should always take into consideration the possibility of associated risks. In the future, we will examine the criteria for selecting patients for whom this method is useful, as well as verify its safety and effectiveness by conducting a prospective randomized analysis. Trial registration number not applicable