The role of hydrosalpinx in IVF: simply mechanical?

I read with interest the manuscript by De Wit et al. (1998) which sheds a new light on the hotly debated subject of the adverse role that hydrosalpinx plays in in-vitro fertilization (IVF). The authors need to be congratulated for successfully showing that women with hydrosalpinges are indeed a heterogeneous group: when hydrosalpinges are clearly visualized by ultrasound, the implantation rate (IR) and pregnancy rate (PR) are markedly reduced compared with other women with hydrosalpinges that were identified by hysterosalpingography or laparoscopy, but not seen on ultrasound. It is unfortunate that in this retrospective study no measurements of the size of the hydrosalpinx were taken, a point also missed in the first study using ultrasound as the diagnostic method for hydrosalpinx (Andersen et al., 1994). Prior investigations, unfortunately again not using size measurements, suggest that women with ‘marked’ distension of one or both Fallopian tubes have a poorer outcome compared to women with ‘slight’ distension (Strandell et al., 1994). However, Andersen et al. (1996), without reporting on measurements, could not find a difference ‘in the largest diameter of the hydrosalpinx between those patients who conceived as compared to those who failed to conceive’. Clearly the impact of the size of the hydrosalpinx needs to be assessed in the ongoing prospective trials to clarify this controversy. While some of the definitions used in De Wit’s manuscript are not shared by most investigators (such as ‘giving a score of one implantation for cases of miscarriages before ultrasonography was performed, in cases of ectopic pregnancy, and when no gestational sac was observed’), the conclusions are critical nonetheless. With more recent publications showing no embryonic toxicity of the hydrosalpinx fluid in humans (Granot et al., 1998; Strandell et al., 1998), unlike earlier studies in mice (Mukherjee et al., 1996; based on which the authors’ recommended ‘prophylactic salpingectomy’), it is clear that the most likely effect of hydrosalpinx is simply mechanical. It is logical to assume that embryonal apposition to the endometrial surface will be compromised when a fluid interface exists. De Wit’s data also indicate that surgical therapy (salpingectomy or proximal tubal interruption) should