OBJECTIVES: The objective of this study was to evaluate the efficacy of cell therapy using human amniotic epithelial stem cells (hAESCs) for the treatment of premature rupture of membranes (PROM) in vitro. DESIGN: Using the amniotic pore culture technique (APCT), we mimicked the environment of PROM in vitro, thus enabling the observation of the healing process of hAESC-treated amniotic membranes. MATERIALS: Amniotic membrane samples were collected from placentas of pregnant women who underwent elective cesarean sections. APCT model and isolated hAESCs were used in this study. All patients who participated in this study provided their written informed consent prior to the commencement of the study. SETTINGS: To create the APCT model in vitro, isolated amniotic membranes were punched to create 5 mm diameter circles and re-punched to form a 1-mm pore at the center. Membranes were cultured in α-minimal essential medium, and the hAESCs were collected and cultured as well. Subsequently, the APCT models were divided into two groups: hAESC treated and control. METHODS: Within the culture period, pore sizes were calculated to evaluate the degree of tissue regeneration in both groups. We then evaluated the histology, cell density, and epithelial thickness of the regenerated tissues. Statistical analyses were performed using SPSS software ver. 20.0 (IBM, Armonk, NY, USA) with repeated-measures one-way analysis of variance or paired samples t test. The significance level was set at p < 0.05. RESULTS: As per the evaluation of the APCT model in vitro, the pore size in the hAESC-treated group reduced by 62.2% on day 6 (62.2 ± 0.19, n = 24), whereas in the control group, it shrank by only 36.8% (p < 0.05) (36.8 ± 0.19, n = 24). Furthermore, the epithelial thickness in the amniotic epithelial stem cell-treated group (10.08 ± 1.26 μm, n = 8) was significantly higher than that in the control group (5.87 ± 0.94 μm, n = 8). Cell density in the regenerated tissue in the amniotic epithelial stem cell-treated group (57 ± 2.77, n = 8) was significantly higher than that in the control group (49 ± 2.23, n = 8). LIMITATIONS: In this study, we did not explore the molecular mechanisms by which hAESCs participate in membrane healing in the APCT model. Although our results showed a significant difference, this difference was not too obvious. Therefore, further research on the mechanisms of hAESCs is needed, with more amniotic tissues and APCT samples being tested. CONCLUSIONS: We developed an APCT model to investigate the PROM conditions in vitro. By implanting donor hAESCs in the pores of the APCT model, we observed that hAESCs seeding accelerated pore healing in vitro. Thus, hAESCs may be a valuable source of cells for cell therapies in regenerative medicine.