Despite recent advancements in reproductive medicine, recurrent implantation failure and habitual abortion remain ongoing issues. One of the most important aspects of successful implantation is the intricate immune response and regulation necessary for the acceptance of the hemiallogenic embryo. The most numerous immune cells in the decidua are uterine natural killer cells (uNK). Studies suggest that changes in the uNK count and physiology may be responsible for the aforementioned pathological conditions. Thus, testing for uNK may provide valuable insights into their pathogenesis. The study compared Pipelle endometrial sampling with conventional curettage to find out whether the less invasive Pipelle method is a viable alternative of tissue collection. Tissue samples from 14 patients obtained by both methods were examined. The average size of tissue samples obtained with Pipelle was 17 mm2, samples obtained with curettage had on average 34 mm2. Using immunohistochemical visualization of CD56 (NK cells) and granzyme B antigens (serine protease-expressing activation state of NK cells), it was found that the average total count of CD56 / mm2 was 115 for Pipelle and 120 for curettage, respectively. The study also proved a correlation between granzyme B positivity and identification of NK cells clusters. The results indicated that Pipelle endometrial sampling seems a suitable method of tissue harvesting for the purpose of uNK cells examination. Pipelle endometrial sampling is safe, cost-effective and can be performed on an outpatient basis without the need of anesthesia or analgesia. Several issues remain yet to be solved: how to standardize the subsequent uNK testing, how to interpret the results and finally yet importantly, how to use this knowledge in personalized treatment protocols.