Revascularization of human ovarian grafts is equally efficient from both sides of the cortex tissue

Research question Does revascularization of human ovarian grafts in a mouse model occur with equal efficiency from both sides of the cortex tissue? Design Twenty-four frozen-thawed ovarian cortex pieces from 12 women were transplanted to immunodeficient mice, for 8 days to analyse graft revascularization using immunohistochemical detection of murine CD31, or for 8 weeks to evaluate follicle density (follicles/mm 3 ). The CD31-positive vessel area and density were quantified using a custom-designed application. Three regions of interest (ROI) were defined in each tissue section: the cortical side, the centre and the medullary side. Vessels were subdivided into three categories according to size: microvessels (<300 µm 2 ), small vessels (300–1000 µm 2 ) and large vessels (>1000–3000 µm 2 ). Results No significant difference in the mean percentage of the CD31-positive vessel area was found between the three ROI (cortical side: 3.9% ± 0.2%; centre: 3.5% ± 0.2%; medullary side: 4.0% ± 0.3%; P = 0.17), but a significantly lower density of vessels was found in the centre of the human ovarian grafts compared with the cortical and medullary sides (cortical side: 323 ± 14 vessels/mm 2 ; centre: 240 ± 12 vessels/mm 2 ; medullary side: 301 ± 18 vessels/mm 2 ; P < 0.001). Microvessels comprised 89–91% of all vessels in the three ROI. Follicle density in ungrafted cortex pieces was 51.8 ± 17.3 and 14.7 ± 3.7 follicles/mm 3 after 8 weeks of xenografting, resulting in a follicle survival rate of 28%. Conclusions Host revascularization was established equally efficiently from both sides of transplanted human ovarian cortex, suggesting that transplantation techniques ensuring revascularization from both sides of the ovarian graft could potentially facilitate faster graft revascularization.