The Best-Fit Model for Endometriosis
Sun-Wei Guo
DOI: https://doi.org/10.1016/s0015-0282(03)00589-2
IF: 6.7
2003-01-01
Fertility and Sterility
Abstract:The pathogenesis of endometriosis remains an enigma in gynecology and a topic of heated debate. After reviewing more than 200 articles on autotransplants and comparison of endometriosis with endometrium, Dr. Redwine (1Redwine D.B. Was Sampson wrong?.Fertil Steril. 2002; 78: 686-693Abstract Full Text Full Text PDF PubMed Scopus (63) Google Scholar) noted that the majority of studies found multiple, profound differences between endometriotic and eutopic endometrium, whereas only a few studies have found similarities between them. He states that “… the balance of evidence suggests that endometriosis is not simply displaced normal endometrium as would be expected in an autotransplant disease.… Either normal native endometrium is profoundly damaged sometime between its eutopic origin and the eventual establishment of disease, or endometriosis does not originate from native endometrium in the first place, but rather has a different provenance.” Finding no evidence supporting the former, he concludes that “endometriosis cannot be considered to be merely displaced normal endometrium. This idea, categorically accepted as one of the mainstays of Sampson’s theory of origin, must be discarded.” Thus, in his opinion, Sampson could be wrong. While Dr. Redwine’s courage to challenge the prevailing view on the pathogenesis of endometriosis is commendable and admirable, his handpicked evidence that leads to his conclusion is flimsy, for the following reasons. First, most, if not all, reported profound differences between endometriosis and eutopic endometrium are post hoc observations—that is, observations made after endometriosis is well established. It is plausible that most of these differences are merely consequences, rather than causes, of the disease endometriosis, as a result of a successful implantation, followed by clonal selections and expansions, and inflammation. Therefore, these somewhat transformed autotransplanted tissues can be quite different from endometrium, especially in a milieu that is quite different from the uterus cavity. Profound as they may be, the differences, by and in themselves, cannot be the ground for concluding that the implantation theory is wrong. Second, Dr. Redwine has overlooked overwhelming epidemiologic evidence and ample clinical evidence resulting from “natural experiments,” that are in support of Sampson’s theory. Several well-designed and well-executed epidemiologic studies reported that shorter length of menstrual cycles, longer duration of flow, heavier menstrual flow, and other menstrual characteristics are risk factors for endometriosis (2Eskenazi B. Warner M.L. Epidemiology of endometriosis.Obstet Gynecol Clin North Am. 1997; 24: 235-258Abstract Full Text Full Text PDF PubMed Scopus (1130) Google Scholar). It has also been well documented that patients with müllerian anomalies and obstructed menstrual flow through the vagina have an increased risk of endometriosis (3Olive D.L. Henderson D.Y. Endometriosis and müllerian anomalies.Obstet Gynecol. 1987; 69: 412-415PubMed Google Scholar). These observations are consistent with Sampson’s theory. Finally Dr. Redwine’s dismissal of the possibility of endometrial tissue damage en route to the peritoneal cavity apparently ignores the third possibility—that is, that some shed endometrial debris may already contain damage, most likely manifested as genomic alterations, that enable this debris to get implanted, grow, cause inflammation, and eventually lead to endometriosis. This possibility is not biologically far-fetched. Certain normal cells in humans spontaneously undergo genomic alterations (4Pettenati M.J. Sweatt A.J. Lantz P. Stanton C.A. Reynolds J. Rao P.N. et al.The human cornea has a high incidence of acquired chromosome abnormalities.Hum Genet. 1997; 101: 26-29Crossref PubMed Scopus (22) Google Scholar). With monthly shedding of the entire endometrium, the cell turnover in endometrium is supposedly fairly high; thus, there are opportunities for disruption of genomic integrity. In addition, endometriotic lesions have been reported to have genomic alterations and loss of heterozygosity (5Gogusev J. Bouquet de Joliniere J. Telvi L. Doussau M. du Manoir S. Stojkoski A. et al.Detection of DNA copy number changes in human endometriosis by comparative genomic hybridization.Hum Genet. 1999; 105: 444-451Crossref PubMed Scopus (84) Google Scholar). This possibility may explain the apparent paradox that while retrograde menstruation is almost universal, only a fraction of women develop endometriosis. We have recently found evidence for genomic alterations in endometria from patients with endometriosis by using microarray-based comparative genomic hybridization (article in preparation). In view of the above, Dr. Redwine’s conclusion may be a little premature. He is certainly correct in saying that “… we in the modern era have intellectual and scientific advantages that have resulted in the compilation of scientific evidence regarding endometriosis that was unavailable to previous generations of clinicians, researchers, and proponents of theories of origin.” But we should not forget that we all stand on the shoulders of the giants of the past. Most reported facts are only true under certain conditions. In the quest for causes of endometriosis, we not only need to gather and sift through a huge body of often conflicting data, but also to “think outside the box.” And in synthesizing new theories while questioning the old, perhaps we also should heed an old adage: Haste makes waste.