Ronald E. Batt1
(1)
State University of New York at Buffalo, Buffalo, New York, USA
Abstract
In a powerful opening statement, Sampson compared the prevalence at operation of “the pathologic conditions arising from the implantations of epithelium which escape from the fallopian tubes into the peritoneal cavity” as second in prevalence only to leiomyomas of the uterus in women between the age of 30 years and menopause.1 Here in the short span of only 2 years after Cullen’s last major publication, Sampson spoke of pelvic adenomas and ovarian hematomas of endometrial type as a common disease entity of women in their fourth and fifth decades. He clarified new terminology used in the title of this paper. “I have discarded the term ‘perforating hemorrhagic cysts’ as applied to this condition, because perforations may occur in other varieties of ovarian hematomas. I now refer to them as hematomas or hemorrhagic cysts of endometrial (müllerian) type.”2
Life History of Ovarian Hematomas of Endometrial Type
In a powerful opening statement, Sampson compared the prevalence at operation of “the pathologic conditions arising from the implantations of epithelium which escape from the fallopian tubes into the peritoneal cavity” as second in prevalence only to leiomyomas of the uterus in women between the age of 30 years and menopause.1 Here in the short span of only 2 years after Cullen’s last major publication, Sampson spoke of pelvic adenomas and ovarian hematomas of endometrial type as a common disease entity of women in their fourth and fifth decades. He clarified new terminology used in the title of this paper. “I have discarded the term ‘perforating hemorrhagic cysts’ as applied to this condition, because perforations may occur in other varieties of ovarian hematomas. I now refer to them as hematomas or hemorrhagic cysts of endometrial (müllerian) type.”2
Sampson addressed the relative virulence of primary and secondary implantations explicitly and in so doing, articulated the active life history, origin, growth, development, and perforation of ovarian hematomas of endometrial type. “The epithelium primarily giving rise to these implantations is derived from or through the fimbriated ends of the fallopian tubes. It lodges either on the surface of the ovaries or on the peritoneal surface of the other pelvic structures, especially those in the culdesac, or on both the ovaries and the pelvic peritoneum, and develops into glands or tubules (adenomas) of endometrial (müllerian) type.3 The primary peritoneal implantation adenomas are usually small and insignificant, but may spread and become invasive. The implantations on the ovary invade tissues of that organ, and as a result of their reaction to menstruation develop into superficial or deep hematomas (hemorrhagic or menstruating cysts) of endometrial (müllerian) type, which usually perforate into the peritoneal cavity. Perforation occurs in the superficial ovarian hematomas while they are still small, a few millimeters in diameter. On the other hand, the hematomas developing in the deeper tissues of the ovary may attain a much larger size, from 1 to 9 cm. in diameter before perforation occurs. The material escaping from the perforation of the ovarian hematoma, whether the latter is small or large, may carry with it epithelium which is cast off from its lining by menstruation. This epithelium may give rise to secondary implantations, which are often apparently more invasive, and have a wider distribution than the primary (original) pelvic implants.”4
In passing, Sampson recognized the ovarian lesions as a possible source of carcinoma.5 However, he regarded the ovary as a nonessential “intermediary host” for the development of implantation adenoma of müllerian type.6 If nonessential, then why did Sampson devote this communication to the life history of ovarian hematomas of endometrial type? He labeled the ovarian “intermediary host” model nonessential in his effort to explain the origin of endometrial (müllerian) type tissue on the surface of the ovary and in ovarian hematomas by his second theory of transtubal retrograde menstruation. Likely, he then realized that his second theory, that of transtubal retrograde menstruation of viable endometrial cells through the narrow and often tortuous lumen of isthmus of the fallopian tubes, would be difficult to prove, let alone easily accepted by the medical profession. Witness Meigs’ enthusiastic acceptance of Sampson first theory, while merely describing Sampson’s second theory as one among several others that attempted to explain the origin of ovarian hematomas of endometrial type.7
I believe Sampson launched his more controversial second theory, the retrograde menstruation and implantation model, to a highly sophisticated audience at the Harvard Medical Society and then let it drift. Meanwhile, he concentrated on gaining acceptance of his first theory of pathogenesis, the ovarian “intermediary host” model. Both theories of implantation rested upon the seed and soil metaphor. “Should the cyst rupture, the hemorrhagic contents, including epithelium and stroma cells, would escape into the peritoneal cavity. Should the epithelium fall on suitable soil, implantation adenoma would arise, which I have designated as secondary implantations, in contradistinction to the primary ones arising from the tube.”8
Sampson was groping for the right terminology. He replaced one awkward and nonspecific term, “perforating hemorrhagic cysts” with the more specific but still imperfect term “hematomas or hemorrhagic cysts of endometrial (müllerian) type.”9 He could not yet decide, which adjective was more correct: endometrial (anatomic term) or müllerian (embryologic term). Sampson explained: “I have used the term ovarian hematomas of endometrial type rather than endometrial hematomas because I believed that in some instances the epithelium lining them may possibly be derived primarily from the tubal mucosa. The term müllerian would be inclusive and a better one.”10
Sampson hinted at the complexity of the pathogenesis of endometriosis when he asked: “Is it possible to decide whether a given implantation adenoma was primarily derived from tubal or uterine mucosa? I do not think it can be definitely proved in any instance but the histologic structure of some of these adenomas suggests a tubal, and others a uterine, origin. Adenomyoma of the uterus and of the tube may be classified as ether primary or secondary. In primary adenomyoma the adenomatous growth is derived from the direct invasion of the uterine or tubal wall by the mucosa lining their cavities. In the secondary type of growth, which in my experience is the much more frequent variety, the uterine or tubal wall is invaded by epithelium implanted on its peritoneal surface, which is derived from the perforation of an ovarian hematoma of endometrial type or from or through the fimbriated extremity of the tube. It is obviously possible that the epithelium giving rise to a secondary adenomyoma of the uterus might be derived primarily either from the tubal or the uterine mucosa, and the same is true of secondary adenomyoma of the tube.”11
Yet, Sampson could not bring himself to use the term müllerian, except bracketed by parentheses. He continued his explication of the life history of ovarian hematomas of endometrial type: “the most natural conception of the source of these tubules is from an abnormal development of the surface epithelium of the ovary [coelomic metaplasia] or from developmentally misplaced epithelium of the müllerian duct [müllerian rests].”12
Ruminating on the idea of developmentally misplaced epithelium, Sampson considered the age of minute adenomas he observed.13 He began questioning the 30 year lower-age boundary. “These hematomas are unusual in women under thirty years of age and if they were developmental and not of acquired origin we would expect them to occur in younger women, soon after puberty.”14 Sampson continued: “They develop during the menstrual life of the patient, when tubal and uterine epithelium would be more likely to escape from or through the fimbriated end of the tube than before puberty and after the menopause.”15 Then, Sampson seemed to contradict himself, allowing the possibility before puberty and after the menopause. “It is possible that the implantation on the ovary of epithelium derived from or through the tube may occur before puberty and after the menopause, and may develop into ovarian cysts or even carcinoma.”16 How can one account for this a seemingly contradictory statement from a careful observer, an observer that Johann Wolfgang von Goethe, Johannes Müller, and Friedrich Wilhelm Karl Heinrich Alexander von Humboldt undoubtedly would embrace? Perhaps, Sampson was attempting to distance himself even further from the embryologic theory of pathogenesis in order to open space for his audience to receive his implantation theory. Realizing the unsettled state of endometriotic theory, Sampson presented a cogent argument to level the field of debate: “it is difficult to disprove any of these theories.”17
Sampson asserted: “We have abundant proof that apparently normal uterine and tubal epithelium may be invasive, as demonstrated in the development of ‘adenomyoma’ from the invasion of the uterine and tubal wall by the epithelium lining these cavities.”18 He described the experiments of a colleague from Albany Medical College and Albany Hospital, the pathologist Jacobson, who conducted the first experiments to test Sampson’s implantation theory.19 Sampson believed that Jacobson had proved experimentally that rabbit uterine mucosa could be transplanted and “give rise to adenoma of endometrial type resembling those described by me except for the absence of any reaction to menstruation; furthermore some of the transplantations made by him in the ovary developed into ‘ovarian cysts.’”20Sampson and his audience were aware that the experiment did not address the issue of transtubal retrograde menstruation, but it did support Sampson’s theory of secondary peritoneal implantation following perforation of an ovarian cyst of endometrial type.
Satisfied that he had presented substantial clinical, pathological, and experimental evidence in support of his first theory of implantation from ruptured ovarian hematomas of endometrial type, Sampson presented circumstantial evidence to support his second theory, that of transtubal retrograde menstruation. He demonstrated that under normal circumstances “the fimbriated end of the [fallopian] tube is frequently situated beneath or lateral to the ovary, with the opening directed towards the surface of that organ by the ‘tether’ like action of the free margin of the mesosalpinx (tubal tether, t.t.) to which the lower fimbriae of the tube are often attached.”21
Sampson described in detail his second theory of pathogenesis of primary ovarian and pelvic adenomas of endometrial (müllerian) type. “This mechanism of the ‘tubal tether’ would facilitate the ovum’s escape into the lumen of the tube in ovulation: this same close anatomical relation between the fimbriated end of the tube and the surface of the ovary would also permit epithelium, escaping from the fimbriated end of the tube or through its lumen, to become implanted on the surface of the ovary. The various physiologic changes in the pelvic contents would also cause the fimbriated end of the tube to brush against the structures in the culdesac. Should implantation adenoma arise from the epithelium escaping from the tube, we should expect to find these implantations most frequently on the lateral and under surface of the ovaries, and on the posterior surface of the lower portions of the broad ligament, the uterus and in the bottom of the culdesac; especially about the uterine attachments of the uterosacral ligaments. It is in these situations that the early implantation adenomas are most often found.”22
Sampson did not include adenomas of endometrial type, which involved the intestinal tract as early lesions, nor did he attribute them to spilled endometrial epithelium from the fallopian tubes. Rather, he believed they “could have arisen from epithelium escaping from a perforated ovarian hematoma.”23 Then, as if he intended to deemphasize the role of the perforated ovarian hematoma and reemphasize the role of the fallopian tube, Sampson recalled explicitly: “On February 14, 1922, before the Harvard Medical Society at Boston, Mass., I reviewed in a general way the entire subject of ovarian hematomas and implantation adenomas of endometrial type. As stated at that meeting my interpretation of the origin and development of these implantations adenomas was as follows: Tubal and uterine epithelium at times escapes into the peritoneal cavity from or through the fimbriated end of the tube.”24
He reviewed data, which he believed supported his second theory of implantation theory. The uterus is often retroflexed; the fallopian tubes “were apparently patent in all” cases;25 he stressed the effect of the tubal tether, the fimbriated end of the tube “is usually found behind the ovary or below and mesial to it26;” and finally Sampson noted that “as women spend the greater portion of the 24 h of the day with the body in the upright posture, whether sitting down or standing, the tendency for the fimbriated end of the tube to be tucked beneath, or lateral to, the ovary would be increased, and ‘sediment’ escaping from the tube would naturally settle on the lateral and the under surface of the ovary and in the bottom of the culdesac, especially its anterior portion.”27 (Italics added)
By explaining that the preferential implantation of shed endometrial and tubal epithelium on the anterior portion of the rectovaginal pouch of Douglas (culdesac) resulted from the combined effects of gravity and female anatomy – the lowest portion of the culdesac being its anterior portion, Sampson provided a forceful and concise explanation for Cullen’s observation that the anterior or retrocervical portion was the very location in which adenomyomas of the so-called rectovaginal septum originated. From that initial retrocervical location, adenomyomas might stay localized, or penetrate anteriorly into the vagina, or grow posteriorly to invade the rectum, or both; in the latter situation resulting in a frozen pelvis with the rectum firmly fused by dense adhesions to the uterus and the pelvic portion of the cervix.
Continuing the image of “upright posture,” Sampson stressed that the distribution of pelvic implantation adenomas of endometrial type arising from the perforation of benign ovarian hematomas corresponded with the distribution of “implantations of carcinoma from the perforation of a malignant ovarian cyst.”28 To complete the correspondence between malignant and benign ovarian cysts and to add emphasis to the extensive pelvic adhesive disease associated with ovarian carcinoma, Sampson made a particularly cogent observation. “All specimens of extensive implantation adenomas which I have studied have been associated with an ovarian hematoma of endometrial type, with evidence of perforation.”29
Still holding the image of “upright posture,” Sampson discussed the distribution of pelvic lesions involving 15 cases formed in the absence of “gross evidence of an ovarian hematoma with perforation.”30His study suggested “that the epithelium from which the implantation arose was derived from the tubal mucosa in some, and from the uterine mucosa (menstruation with a back flow through the tube) in others.”31 Sampson continued: “It was interesting to note the character of the implantations when there was no gross evidence of an ovarian hematoma with perforation. They were usually smaller, less invasive, and not as widely distributed as those generally found in the pelvis associated with an ovarian hematoma with evidence of perforation.”32 Not only did Sampson discern a difference in the distribution pattern, he also observed a histological difference. “Many of them also presented a little different histologic picture; they usually did not resemble typical endometrium as closely as did the implantations which were associated with an ovarian hematoma with perforation.”33
Recall that this, his third case series, represented early stages of the disease.34 Sampson was taking great pains to describe the appearance and distribution of peritoneal and ovarian adenomas of endometrial and tubal type and he needed to see these implants in the early stage of their life history. Did Sampson use a proctoscope inserted through a small incision in the abdomen to detect early implants in the rectovaginal pouch of Douglas? In other words, instead of observing through a large laparotomy incision, did Sampson use a large-bore proctoscope to examine the pelvis in patients he suspected had early implantation adenomas of endometrial and tubal type, as gynecologists would use a laparoscope later in the twentieth century? I suspect he may have done so. By inserting a proctoscope – a round rectal speculum – through a small incision in the lower abdomen and then by placing the patient in head down steep Trendelenburg position, a position that would allow the intestines to slide up toward the diaphragms, Sampson could have obtained a clear visualization of the pelvis. But to attribute such foresight to Sampson is speculative and anachronistic without evidence that Sampson used a large-bore proctoscope.
In fact, there is indirect circumstantial evidence. While a resident at Johns Hopkins in 1903, Sampson published a method of controlling hemorrhage following pelvic surgery by packing the pelvis with gauze inserted through a proctoscope and maintaining counter pressure by packing the vagina with gauze.35 The caption under Figure 3 of that paper reads: “Figure 3. Introduction of proctoscope into pelvis. The lower end of the abdominal incision has been opened, and the proctoscope has been inserted through the opening into the abdominal cavity, taking care to direct the end well forward against the bladder, so that the coils of the intestines will be displaced backwards. In this instance the uterus with appendages, have been removed leaving in the cervix. By removing the obturator the various portions of the pelvis may be inspected, and here one can see that there is a small accumulation of fluid in the pelvis.”36
Fitzgerald wrote the historical note, with the quotation cited above, on the twentieth anniversary of Sampson’s death. Of all of Sampson’s publications, he cited only six, four related to cervical cancer, one to endometriosis, and the one concerning pelvic packing through a proctoscope. This would indicate to this writer that the proctoscopic packing procedure had been used in Albany since 1905 when Sampson became Professor of Gynecology and possibly was still used 20 years after Sampson’s death. It is not difficult to imagine Sampson directing reflected light from a head-mirror, down the short but large-bore proctoscope to visualize early implants on the ovaries and pelvic peritoneum. Howard Kelly had devised such a head-mirror in 1883 for use in gynecology.37
Sampson returned to the life history of ovarian hematomas of endometrial type. He addressed the development and regression of superficial and deep ovarian hematomas of endometrial type, noting that “the early stages of development of the deeper hematomas are the same as the early stages in the development of superficial hematomas.”38 He postulated that in superficial hematomas, “if all the epithelial lining is cast off by menstruation the life of the hemorrhagic cyst is ended, and all evidence of it may disappear.”39 With respect to deep ovarian hematomas of endometrial type, Sampson stated that “it was impossible in every instance to determine the source of the tubules from which the hematomas arose, but in many specimens it was evident that the hematoma had developed from a gland or tubule which had invaded the ovarian tissue from its surface…enough sections were made from some specimens to demonstrate that multiple small hematomas may arise in the tortuous course of a tubule, just as multiple ponds and lakes may arise in the course of a stream.”40 The life history of a deep ovarian hematoma may be prolonged by repeated rupture into the peritoneal cavity,41 or it may grow so slowly that it remains quietly within the ovary without perforating and regresses after the menopause, 42 and lastly, “Some of these hemorrhagic cysts or hematomas lose all of their epithelial lining and ‘die’ before they reach the surface of the ovary and perforate.”43
Sampson had only one opportunity to observe the effects of pregnancy on an ovarian hematoma of endometrial type. At hysterectomy and removal of both ovaries for uterine fibroids in a 37-year-old woman, he opened the uterus and found an early pregnancy.44 From this, it was obvious that the corpus luteum in the right ovary was a corpus luteum of pregnancy. He found “the hematoma of the left ovary was lined by typical decidual tissue, with the surface epithelium still present in the depressions. Histologically it was identical with that of the compact layer in the decidua vera of the pregnant uterus. Glands were not present in the lining of the ovarian hematoma.”45 Recall that while Sampson was still a resident at Johns Hopkins, J. Whitridge Williams had demonstrated decidual reaction in an adenomatous uterus at the autopsy of a woman who died shortly after delivery.46 In Williams’ case, the decidual reaction occurred in internal (uterine) endometriosis, whereas in Sampson’s case, decidual reaction was observed in external (ovarian) endometriosis.
Once again Sampson addressed the differing ages of the small superficial lesions; indicating the implants represented different generations, indicative of the chronicity of the disease process. “If they have recently reacted to menstruation they are red in color, later bluish black. They are usually multiple, and are often found in various stages of development and retrogression in the same specimen, suggesting that they are not all of the same age.”47This observation is perfectly compatible with his theory of repeated retrograde menstruation and implantation producing different generations of peritoneal disease.
Lastly, Sampson addressed the influence of menopause and old age on “ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type.” He based his comments on the two cases he had seen. In the first case, he found “caked” menstrual blood within the cavity of an ovarian hematoma “lined by a wavy hyaline membrane deeply pigmented with hemosiderin. The epithelial lining had completely disappeared.”48 The blood serum within the ovarian hematoma had been absorbed leaving behind the more solid “caked” residue. However, in the same menopausal women, he observed “implantation adenoma was present invading the posterior uterine wall, which histologically resembled the mucosa of the uterine cavity.”49 Sampson had not had time to thoroughly study the second case, but he did comment that an “epithelial lining was present” in two ovarian hematomas in a 59-year-old woman with endometrial carcinoma.50
Sampson presented the following evidence that the “hematomas or hemorrhagic cysts [were] of endometrial (müllerian) type.”51 First, he argued similar structure and function: “they develop from glands or tubules in the ovary which are lined by cuboidal or columnar epithelium (often ciliated) resembling tubal and uterine epithelium. Hemorrhage occurs in the ovarian tissue about the glands or tubules at the time of menstruation.”52 Second, he argued for histological similarity between these secondary ovarian lesions and primary adenomyomas of the uterus and fallopian tubes: “Histologically, the epithelial lining of the ovarian hematomas is similar to that of the hematomas and dilated cavities found in primary adenomyoma of the uterus and of the tube…Every variation in the histologic structure of the lining of the ovarian hematomas (often seen in different portions of the same hematoma) is due to different phases in its reaction to menstruation.”53 He concluded that “perforation of the hematoma is but a result of menstruation…but their reaction to menstruation, pregnancy, and old age is similar to that of the mucosa of the uterine cavity.”54
Before he presented the 20 case reports, Sampson reminded his readers that his first case series published in 1921 and the second published in 1922, for the most part, represented “long standing [or] advanced stages of the disease.”55 In the 1921 case series, all the ovarian hematomas had evidence of perforation.56 His second case series, published in 1922, included cases with implantation adenoma of endometrial type involving the intestinal tract.57This, his third case series, represented early stages of the disease. By early stage, Sampson meant an early stage of ovarian hematomas of endometrial type; ovarian hematomas that had not perforated. Sampson stated: “The cases reported in the present [third] series were chosen from those demonstrating the origin and development of ovarian hematomas of endometrial type and ovarian hematomas without perforation rather than from those with perforation and extensive implantations.”58
However, on close examination of the cases, it is evident that, while they may well have represented the earliest cases Sampson could gather by 1922, they were not sufficient for his purposes. In cases 1, 6, 7, 9, and 14, the ovaries had perforated. In cases 10 and 11, Sampson indicated “? perforated.”59 In other instances such as cases 4, 5, 8, 13, 16, and 17, Sampson used such qualifiers as “probably arose from…could have arisen from…may have arisen from…suggested that…could have had a common origin.”60 In only four cases, cases 12, 15, 18 and 19, did Sampson have material that substantiated the opening statement in his conclusion: “Next to leiomyomas of the uterus, the pathologic conditions arising from the implantation of epithelium which escapes from the fallopian tubes into the peritoneal cavity probably furnish the most frequent pelvis lesions found in woman between the ages of 30 and the menopause.”61 I believe Sampson realized he needed more evidence. To his concluding rhetorical query of whether “the primary ovarian and peritoneal implantation (those developing from epithelium escaping from the fallopian tube) arise from both tubal and uterine epithelium,” Sampson answered “the specimens which I have studied would suggest that they may.”62 He answered in the passive voice, and the operative word in his answer was “suggest.” Nonetheless, he seems to have been convinced, as witnessed by a preview of work in progress. “I believe that the implantation adenomas in the ovary derived from tubal and uterine epithelium are the source of many ovarian cysts and carcinomas, and am convinced that two of the latter, which I am studying at the present time, arose from this source.”63
In sum, Sampson had two theories of pathogenesis of implantation endometriosis: the ovarian and the tubal theories. He thought the tubal theory was in a preliminary state in 1922 and had to be presented guardedly even to as sophisticated an audience as the American Gynecological Society. There were still a lot of unknowns and Sampson’s terminology reflected his state of partial understanding as he emphasized “The Life History of Ovarian Hematomas (Hemorrhagic Cysts) of Endometrial (Müllerian) Type.”
President’s Address: An Autobiographical Essay
John Albertson Sampson had been elected by unanimous vote to membership in the prestigious and excusive American Gynecological Society in 1906, 1 year after completing his residency in Gynecology at Johns Hopkins Hospital and 1 year after his appointment as Clinical Professor of Gynecology at Albany Medical College. Cullen, then emeritus Professor of Gynecology at Johns Hopkins, recorded in an obituary that “a letter has been preserved from Dr. Howard A. Kelly saying this [Sampson’s case] is the only case where anyone ever received all the votes.”64 Such was the popularity and academic prowess of Sampson in 1906, already the author of 17 papers on gynecologic subjects.
On the occasion of his election to the presidency of the American Gynecological Society in 1923, Sampson chose to honor Theodore Roosevelt and Howard A. Kelly, role models of his youth and professional life, respectively.65 In essence, this was an autobiographical essay that revealed the formative elements of Sampson; the botanist, the biologist, the surgeon, the investigator, the professor.
From Roosevelt, his youthful hero, he learned the importance of two hobbies as formative for a career in medicine, reading and the study of natural history. As a boy, Sampson, like Roosevelt, studied ornithology.66 Both learned to observe and record observations, to apply themselves in mastering the subject and to draw their own conclusions. Sampson enumerated four qualities that he believed were largely responsible for Roosevelt’s success in life: “first, his personality; second, his energy and persistence; third, the training in learning how to acquire knowledge derived in large part from his pursuit of ornithology; and fourth, his ability to impress people with his thoughts by his convincing speaking and forceful writing, i.e., his ability to use the English language. The hobbies of Roosevelt were of great value to him as a statesman; the same ones would likewise be of great value to a physician, for the habits and methods of observation, of recording observations, and arriving at a diagnosis in the study of ornithology could easily be applied to the study of medicine, both ornithology and medicine being branches of biology. Physicians, as well as statesmen, should know how to use the English language.”67
Sampson recognized a fellow naturalist when he encountered Howard Kelly in medical school.68 From Kelly the biologist, herpetologist, and consummate pelvic and abdominal surgeon, Sampson learned the fundamentals and fine points of his lifelong professional hobby, investigation of the malignant invasive disease – cervical cancer, and the benign invasive disease – endometriosis. Like Kelly, Sampson loved the out-of-doors. Whereas Kelly vacationed with Cullen every summer on the rivers and lakes in the bush country of Ontario, Canada; Sampson spent his leisure feeding wild animals and enjoying his woods, pond, and Adirondack-style Hawthorne Lodge in Grafton, New York. Employing a biological analogy, Sampson said: “Biology is the study of living beings, both plant and animal, including man, and medicine is but one of its many branches…There are two varieties of naturalists, the collector and the investigator. One answers the question what and the other the question why.69 The botanist who collects answers the question what – and can always remain a botanist, but the one who studies the life history of plants and their struggle for existence, who answers the question why, must become a naturalist in the broadest sense of the word.”70
Sampson spoke of premedical education and the basic sciences so important to the education of physicians: anatomy, physiology, chemistry, and pathology. “They are all indispensable for a better foundation in the study, diagnosis, treatment and prevention of disease, but they are inseparable from each other and from clinical medicine.71 Of these basic sciences pathology is the most important, as it furnishes a definite understanding for the clinical study of disease, namely living pathology; and the two are inseparable.”72 Sampson then elucidated the theme of “living pathology” that guided his scientific research as it had that of Howard Kelly. “The clinician, especially the surgeon, has a wonderful opportunity to study living pathology, which the laboratory worker unfortunately rarely sees, except in experimental work on the lower animals. To do their best work and make their greatest contributions, clinicians should be fundamentally and eternally pathologists, and pathologists should, at least occasionally, be clinicians.”73
By stressing the clinical aspects of disease and living pathology that naturally led to explanatory theory and therapeutic innovation,74 Sampson reversed the reciprocal relationship between laboratory and clinic that Rokitansky had emphasized in mid-nineteenth century. The Danish physician and nosographers, Knud Faber, expressed Rokitansky’s position clearly. “Rokitansky…in his textbook of 1846, drew the oft-cited conclusion ‘that pathological anatomy should be the base not only of the knowledge of physicians but also of their practice, as it contains all there is in medicine of positive knowledge and the foundation of it!’ With such views scientists had wandered far from the nosography of the English [Sydenham] and French [Bichat and Laennec], and landed in anatomical diagnosis of the purest and driest sort. Is it to be wondered that it brought with it a complete therapeutic nihilism in internal medicine.”75
Having revealed his modus operandi, we can better appreciate Sampson’s careful observations expressed in the title of his classic contribution to the pathogenesis of pelvic endometriosis:76Peritoneal endometriosis due to the menstrual dissemination of endometrial tissue into the peritoneal cavity, published in 1927.77
Footnotes
1
Sampson, JA. The life history of ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type. American Journal of Obstetrics and Gynecology 1922;4:451–512:451.
2
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:452:454.
3
Note how unsettled and awkward the terminology at this juncture: glands or tubules (adenomas) of endometrial (müllerian) type.
4
Sampson, JA. The life history of ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type. American Journal of Obstetrics and Gynecology 1922;4:451–512:451–452. Sampson raised the perennial question, how does the surgeon distinguish, which lesions are more likely to become invasive?
5
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:452.
6
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:452.
7
Meigs, JV. Endometrial hematomas of the ovary. Boston Med Surg J 1922:187:1–13.
8
Sampson, JA. The life history of ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type. American Journal of Obstetrics and Gynecology 1922;4:451–512:474. See description accompanying Fig. 29. - Plate III (Case 5) on the bottom of page 474. See also the text on page 478, which reads: When “perforation [of the ovarian cyst] occurs and some of the contents of the cyst escapes into the peritoneal cavity carrying with it epithelial cells which apparently give rise to implantation adenomas wherever the epithelium falls on suitable ‘soil.’”
9
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:454. “I have discarded the term “perforating hemorrhagic cysts” as applied to this condition, because the perforations may occur in other varieties of ovarian hematomas. I now refer to them as hematomas or hemorrhagic cysts of endometrial (müllerian) type. Their epithelial lining, where present, is similar to that found in the hematomas due to the retention of “menstrual” blood, which occur in the adenomyomas of the uterus derived from the uterine mucosa, and the blood in the ovarian hematomas is also apparently of menstrual origin.”
10
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:457.
11
Sampson, JA. The life history of ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type. American Journal of Obstetrics and Gynecology 1922;4:451–512:457–458.
12
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:461.
13
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:462. Ibid: 468, Fig. 16. (Case 16). “Photomicrographs…of the adenomas of endometrial (müllerian) type invading the lateral surface of the right ovary and posterior surface of the uterus…Their histological structure is similar, and I believe that they both have a common origin from epithelium escaping from or through the fallopian tube. It is impossible to decide, from their structure, whether the epithelium was derived from the tubal or the uterine mucosa.”
14
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:462.
15
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:463.
16
Sampson, JA. The life history of ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type. American Journal of Obstetrics and Gynecology 1922;4:451–512:463.
17
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:462.
18
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:465.
19
Jacobson VC. The autotransplantation of endometrial tissue in the rabbit. Archives Surgery 1922;5:281–300.
20
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:465–466.
21
Sampson, JA. The life history of ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type. American Journal of Obstetrics and Gynecology 1922;4:451–512:453. Quotation from the legend under Figure 1.
22
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:453. Quotation from the legend under Figure 1.
23
Sampson, JA. The life history of ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type. American Journal of Obstetrics and Gynecology 1922;4:451–512:455. See also Sampson JA. Intestinal adenomas of endometrial type: their importance and their relation to ovarian hematomas of endometrial type (perforating hemorrhagic cysts of the ovary). Archives of Surgery 1922;5:217–280.
24
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:456.
25
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:463.
26
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:466.
27
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:467.
28
Sampson, JA. The life history of ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type. American Journal of Obstetrics and Gynecology 1922;4:451–512:468.
29
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:469.
30
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:469.
31
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:469–70.
32
Sampson, JA. The life history of ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type. American Journal of Obstetrics and Gynecology 1922;4:451–512:469.
33
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:469.
34
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:485. By early stage, Sampson meant an early stage of ovarian hematomas of endometrial type, ovarian hematomas that had not perforated. Specifically Sampson stated: “The cases reported in the present [third] series were chosen from those demonstrating the origin and development of ovarian hematomas of endometrial type and ovarian hematomas without perforation rather than from those with perforation and extensive implantations.”
35
Sampson JA. Control of hemorrhage following pelvic operations by packing the pelvis with gauze through a proctoscope and maintaining counter pressure by packing the vagina. Johns Hopkins Hosp Bull 1903;14:237–42.
36
Fitzgerald WJ. John Albertson Sampson, MD: pioneer gynecologist, teacher, and researcher. NY State J Med 1966;66:1244–7: 1246. Dr. Fitzgerald was a clinical instructor in the Department of Obstetrics and Gynecology, The Albany Medical College of Union University at the time of Sampson’s death.
37
Kelly HA. An improved attachment for the head-mirror. Med News Philadelphia 1883:xliii:390.
38
Sampson, JA. The life history of ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type. American Journal of Obstetrics and Gynecology 1922;4:451–512:473.
39
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:472.
40
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:473.
41
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:478. Sampson believed “the escape of the contents of the cyst probably prolongs its life. It relieves tension and thus favors repair until the next reaction to menstruation.” See page 510. “The hematomas developing in the deeper tissues of the ovary may attain a large size, several centimeters in diameter, before perforation occurs. As the menstrual blood is retained in the cavity of the hemorrhagic cyst and in the stroma of the lining for a long time, many interesting histologic changes occur in the wall of the cyst in the attempt to absorb the menstrual blood, and to reline the denuded surface by epithelium from that which had not been removed by menstruation. The development and activities of the endothelial leucocytes, which act as scavengers, play an important part in the absorption of the menstrual blood and the deposit of the pigment, derived from this blood, in the walls of the hematoma. Perforation permits the contents of the hematomas to escape into the peritoneal cavity, and may temporarily relieve the embarrassment caused by its retention. The perforation is sealed by the ovary or cyst becoming adherent to adjacent structures at the site of its perforation. The hematoma again fills up with blood at its next reaction to menstruation, and repeated perforations may occur. As the reaction to menstruation is destructive, and as the repair and regeneration of the epithelial lining is accomplished under great difficulties (due to the retention of the menstrual blood), the ultimate tendency of the hemorrhagic cyst is one of retrogression.”
42
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:473. Sampson described the mechanism in detail on page 473–474.
43
Sampson, JA. The life history of ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type. American Journal of Obstetrics and Gynecology 1922;4:451–512:475.
44
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:504. See Fig. 66.-(Case 14) on page 504 entitled: Sagittal section of the uterus shown in the preceding illustration, demonstrating the early pregnancy (embryo 14 mm long) (x 3/5). You might ask, how could Sampson miss diagnosing this early pregnancy and avoid the hysterectomy? After all he describes the 37 year old woman as sterile and having wanted children. The answer is two-fold. First this was before pregnancy tests were available, moreover her periods or uterine bleeding was regular. Secondly, from examination of the illustration, there was a large intramural fibroid anteriorly, a smaller subserous fibroid protruding as a knob from the top of the uterus, and a large fibroid protruding from the posterior uterus. From my own clinical experience, the strategic position of the fibroids, which by their name reveals their fibrous nature, precludes any appreciation of softening of the underlying uterus associated with pregnancy.
45
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:481-482.
46
J. Whitridge Williams. Decidual formation through the uterine muscularis: a contribution to the origin of adenomyoma of the uterus. Transactions of the Southern Surgical Association 1904;17:119.
47
Sampson, JA. The life history of ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type. American Journal of Obstetrics and Gynecology 1922;4:451–512:472.
48
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:583.
49
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:483.
50
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:483.
51
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:483.
52
Sampson, JA. The life history of ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type. American Journal of Obstetrics and Gynecology 1922;4:451–512:483.
53
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:483.
54
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:483–484.
55
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:485.
56
Sampson JA. Perforating hemorrhagic (chocolate) cysts of the ovary: their importance and especially their relation to pelvic adenomas of endometrial type (“Adenomyoma” of the uterus, rectovaginal septum, sigmoid, etc.) Archives of Surgery 1921;3:245–323.
57
Sampson JA. Intestinal adenomas of endometrial type: their importance and their relation to ovarian hematomas of endometrial type (perforating hemorrhagic cysts of the ovary). Archives of Surgery 1922;5:217–280.
58
Sampson, JA. The life history of ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type. American Journal of Obstetrics and Gynecology 1922;4:451–512:485.
59
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:495–496.
60
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:488, 489, 493, 500, 503, 505.
61
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:509. See pages 497–498. Case 12: 41-year-old woman with “bilateral hematosalpinx and ‘adenomyoma’ of the distal ends of both tubes with extension through to their peritoneal surface, implantation adenoma of endometrial type on the posterior surface of the uterus, small intramural leiomyomas of the uterus (no evidence found of hematomas of endometrial type in the ovaries).” Sampson believed “that the adenoma on the posterior surface of the uterus arose from this source…miniature endometrial hematomas” in the distal end of both fallopian tubes.
See pages 501–502. Case 15: 36-year-old woman with implantation adenoma of endometrial type on the mesial surface of the right ovary and on the anterior surface of the uterus just below the attachment of the right round ligament. Sampson believed “that the implantations on the ovary and on the anterior surface of the uterus probably arose from epithelium escaping from the tube.”
See page 506–507. Case 18: 22-year-old woman with implantation adenoma (of endometrial type) of the posterior uterine wall, culdesac, and posterior surface of the right broad ligament; retroflexion of the uterus. Sampson noted: “the uterus was found to be retroflexed and after replacing it a pigmented (hemorrhagic) elevation about 2 mm. in diameter was noticed on the posterior surface of the right uterine cornu, and a similar but broader elevation (implantation) was detected in the culdesac, at place which exactly came in contact with the implantation on the uterus when the latter was replaced in retroflexion. A similar implantation (but not hemorrhagic) was found on the posterior surface of the right broad ligament. Both tubes and ovaries appeared normal, and the latter were examined very carefully for implantation…These implantations were in situations which at times could have been in contact with the fimbriated end of the right fallopian tube, and I believe arose from epithelium escaping from the tube. The implantation on the surface of the uterus might have been a contact implantation from the one in the culdesac, or vise versa.”
See page 507–509. Case 19: 32-year-old woman with implantation adenoma (of endometrial type) of the mesial surface of the right ovary, the right broad ligament between the tube and the ovary and on the suspensory ligament of the ovary; retroflexion of the uterus. “The situation of the implantation, especially the one directly in front of the fimbriated end of the tube, on the suspensory ligament of the ovary, would indicate that they arose from epithelium escaping from or through the fimbriated end of the tube.”
62
Sampson, JA. The life history of ovarian hematomas (hemorrhagic cysts) of endometrial (müllerian) type. American Journal of Obstetrics and Gynecology 1922;4:451–512:511.
63
Sampson, JA. American Journal of Obstetrics and Gynecology 1922;4:451–512:512.
64
Cullen TS. In Memoriam: John Albertson Sampson: 1873–1946. Trans Am Gynecol Society 1947:70:273–4.
65
Cullen TS. In Memoriam. Trans Am Gynecol Society 1947:70:273–4. “Especially to Dr. Howard Kelly was Dr. Sampson indebted for the stimulating example of a great surgeon and God-loving man.”
66
Cullen TS. In Memoriam Trans Am Gynecol Society 1947:70:273–4. In 1890, at the age 17, Sampson “wrote a report on the evening grosbeak for Forest and Stream.”
67
Sampson JA. President’s Address. Fundamental elements in the advancement of medicine. Am J Obstet Gynecol 1923;6:1–11:2–3.
68
Sampson JA. President’s Address. Am J Obstet Gynecol 1923;6:1–11:4, Kelly had a lifelong interest in botany and from this interest wrote a book entitled “Some American Medical Botanists.” Kelly was also interested in poisonous snakes. Kelly HA. Poisonous snakes. Johns Hopkins Hosp Bull 1900;xi:73.
69
Sampson JA. President’s Address. Am J Obstet Gynecol 1923;6:1–11:5.
70
Sampson JA. President’s Address. Am J Obstet Gynecol 1923;6:1–11:4.
71
Sampson JA. President’s Address. Am J Obstet Gynecol 1923;6:1–11:8–9.
72
Sampson JA. President’s Address. Am J Obstet Gynecol 1923;6:1–11:11.
73
Sampson JA. President’s Address. Am J Obstet Gynecol 1923;6:1–11:9.
74
T.K.A.B. Eskes and L.D. Longo, eds. Classics in Obstetrics and Gynecology: Innovative Papers that Have Contributed to Current Clinical Practice [Pearl River, NY: Parthenon Publishing Group, 1994], 277–236.
75
Knud Faber, Nosography in Modern Internal Medicine [New York: Paul B. Hoeber, Inc., 1923], 57.
76
T.K.A.B. Eskes and L.D. Longo, eds. Classics in Obstetrics and Gynecology: Innovative Papers that Have Contributed to Current Clinical Practice [Pearl River, NY: Parthenon Publishing Group, 1994], 277–236. The editors reprinted Sampson’s paper. Editorial comments followed on pp. 327–337.
77
Peritoneal endometriosis due to the menstrual dissemination of endometrial tissue into the peritoneal cavity. Am J Obstet Gynecol 1927;l4:422–469.