Endometrial cell dissemination at diagnostic hysteroscopy: a prospective randomized cross-over comparison of normal saline and carbon dioxide uterine distension.

The incidence of pelvic spreading of endometrial cells at diagnostic hysteroscopy was studied comparing the two distension media carbon dioxide (CO(2)) and normal saline (N/Saline). Thirty patients requiring laparoscopy and hysteroscopy were included in this study, the main indication for surgery being subfertility. Hysteroscopy was performed using both CO(2) and N/Saline distension on each patient, the order of the distension media being randomly allocated. Samples of peritoneal fluid were aspirated from the pouch of Douglas before and after hysteroscopy with each distension medium, and the specimens were investigated cytologically for the presence of endometrial cells. Endometrium was present in 2/30 (6.7%) peritoneal aspirates before and in 15/60 (25%) collected after the hysteroscopies. There was no major difference between liquid or gaseous distension, transtubal reflux of endometrial cells occurring in 7/30 (23.3%) and in 8/30 (26.7%) hysteroscopies respectively. Positive peritoneal cytology was observed significantly more often in patients who were in the proliferative phase of the menstrual cycle [9/14 (64.3%) versus 0/11, P < 0.004]. In conclusion, transtubal dissemination of endometrium occurs in about one quarter of patients, irrespective whether N/Saline or CO(2) is used for uterine distension; there is no advantage to using gaseous distension for hysteroscopy when investigating high-risk cases for endometrial malignancy.