Paper - Tubal gestation - A pathological study

From Embryology
Embryology - 22 Apr 2024    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Berkeley C. and Bonney V. Tubal Gestation - A Pathological Study. (1905) Brit. J. Obst. and Gyn. 7(2): 78-96.

Online Editor  
Mark Hill.jpg
This paper by Berkeley and Bonney are early descriptions of tubal (ectopic) pregnancy.

Modern Links: ectopic pregnancy

Ectopic Links: ectopic pregnancy | implantation | Week 2 | placenta abnormalities | trophoblast | uterus | Chlamydia | Ultrasound - Ectopic Pregnancy | ultrasound | Ectopic Implantation Research
Historic Ectopic Pregnancy  
1888 Ectopic pregnancy | 1895 Ectopic pregnancy | 1914 Infundibulum implantation | 1915 Tubal gestation | Historic 1920 Paper | 1921 Extra-uterine pregnancy | 1922 Tubal pregnancy1934 Early ectopic
Carnegie Tubal Embryos: 109 | 1390 | 1544 | 175 | 256 | 389 | 390 | 3956 | 4160 | 422 | 431 | 4476 | 5048 | 5206 | 576 | 6097 | 6150 | 6551 | 657 | 6631 | 6686 | 670 | 6750 | 6909 | 6950 | 7433 | 7897 | 7900 | 8005 | 8098 | 8147 | 8179 | 8235 | 8355 | 899 | 966
Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic" (textbooks, papers, people, recommendations) appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms, interpretations and recommendations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

Tubal Gestation: A Pathological Study


Comyns Berkeley, B.A., M.B., B.C. (Cantab), M.R.C.P.,

Lecturer on Midwifery and Assistant Obstetric Physician to the Middlesea: Hospital; Senior Physicia/n. to Out-Patients, Chelsea Hospital for Women.


Vlcron Bonney, M.S., M.D., B.Sc. (Lond.), F.R.C.S., M.R.C.P.,

Lecturer on Practical Midwifery and Obstetric Tutor to the Middlesez Hospital; Physician to Out-Patients, Chelsea Hospital for Women.

(From the Clinical Reaearch Department of the Middlesex Hospital).


THE first recorded case of extra-uterine gestation occurs in the writing of Albucasis in the eleventh century. The author states that an abscess formed at the umbilicus, from which foetal bones extruded, in a patient in whose uterus the fetus had died :-

Ego quondam vidi mulierem, qua: grsvida fuerat, fcetus in utero ejus moriebatur, deinde alia insuper vice gravida facts est deinde moriebatur foetas ille alter etism et accedit illi post longam tempus tumor in umbilico ejus qui inflatus fuit donec aperuit et pus produxit. Ex loco egreditur os; deinde prseterierunt dies aliquot et egreditur os slterum. Ego igitur opinabar, haze esse ex ossibus fuatus mortui. Vulnus igiter investigavi, et ex illo ossa multa. extraxi; mulier autem optime se habuit, adeoque vixit illo modo longum tempus, pauco pure ex illo locoproducti.”— Lib. ii., sec. 74.

The next case would appear to be recorded by Polinus, 1531, but will not bear critical investigation. The patient at her tenth pregnancy developed an abscess in her left hypochondrium, which bursting permitted the exit of a living male child; the mother died on the third day in great agony.

In the early half of the sixteenth century Cornax dilated an ulcer near the umbilicus, and extracted therefrom a semi-putrid foetus which had been retained for five years. (“Suc. Ess. Hist., sur les accouch.,” Vol. ii., p. 61.)

A case is mentioned by Felix Platerus occurring in the concubine of a member of one of the sacerdotal orders, in which, after labour pains had continued for eight days and then subsided without delivery, a swelling appeared at the umbilicus from which a dead foetus was extracted. (Felix Platerus “ De partium corporis,” etc., 1597.)

Cordaeus, who lived at the same time as Platerus, records a case where a patient remained unfruitful till she was 40 years of age. She then became pregnant, but labour pains passed off without a child being born; at the age of 71 she died, and a lithopaadion was found in her abdomen. (Cordax “Comment. in librum priorem Hippocrat de Mulieribus,” p. 740.)

Horstius published in 1563 a case of spurious labour occurring in 1549. The patient still retained the foetus at the date of his writing. (Horstius oper., Med. Lib. xi., p. 504.)

Primrose in 1595 (Primrosii Morb., Mul. Lib. iv., p. 316) and Hildanus (Fabricius Hildanus, p. 908) reported similar cases.

Riolanus (Riolan. Fil. anthropogr., Lib. ii.) was the first to definitely describe a case in which the foetus was found in the Fallopian tube. This occurred in 1604, and he recorded 9. similar case in 1638.

Since this many cases have been reported from time to time, but even up to 1876 the condition was apparently thought to be a very rare one. Hennig (“Die Krankheiten der Eileiter und die Tuben— schwangerschaft. Stuttgart,” 1876) stated that even the directors of large obstetrical institutions might never meet with a case, and Parry in 1876 could only collect 500 cases from the literature dealing with the subject up to that time.

It remained for Lawson Tait in 1883 to expand our knowledge of this subject by operating on a case of tubal rupture, thus bringing the condition within the domain of abdominal surgery.

Hermann in 1890 first diagnosed a case of tubal pregnancy before rupture and operated upon it. Since then very great advances have been made in our knowledge, not only in diagnosis and treatment, but also as regards the pathology of the condition.

Up to 1898 the developing ovum was held to be situated in the lumen of the tube, but since that date the work published by many observers on the method of implantation of the intra-uterine ovum, together with more accurate microscopical study of the pregnant tube, has led to the discovery that the tubal ovum is situated, not in the lumen of the Fallopian tube, but in a sac in the wall of the tube.

Fiirth first described in 1898 a two-and-a-half weeks’ tubal gestation in which the developing ovum lay entirely outside the lumen of the tube, being separated from it by a thin layer of muscular tissue, but the occurrence was not then recognised as being a normal one.

Hubrecht in 1889 published a monograph on the placentation of the hedgehog, showing that the developing ovum in this animal, after sequestration in a crypt, becomes imbedded in the sub-epithelial portion of the mucoa of the uterus by the action of the non-foetal ectodermal cells of the blastocyst. These cells Hubrecht named Trophoblast because of their nutritive function, for by their agency the blastocyst “burrows” into the maternal tissues, both destroying and absorbing them.

Similar observations have subsequently been made on bats (Duval), the guinea-pig (V. Spee), the rat and mouse (A. Robinson, Sobotta), all establishing the fact that in these animals the non-ftetal ectoderm of the blastocyst destroys the uterine epithelium, and subsequently the sub-epithelial portion of the mucosa, until the blastocyst comes to lie in a space (implantation space) within the maternal tissue.

In 1899 a very early human ovum (5-7 days) was described by Peters, which demonstrated very clearly that what applied to the ovum of rodents and insectivora was equally applicable to man, viz., that the human ovum was an imbedded one.

Up to this time the intra-uterine implantation of the human ovum had been held to be efiected by a surface interlocking of fcetal and maternal tissues similar to that obtaining in the placenta of ungulates such as the sheep or sow, and the formation of the decidua reflexa had been explained by the upgrowing of mucosal folds around the superficially attached ovum.

“The human blastocyst described by Peters was completely buried in the uterine submucosa.

“The wall of the blastocyst was at its outer part composed of a manylayered cell mass constituting the trophic ectoderm or trophoblast, which was irregularly excavated by blood extravasations from the maternal vessels. The maternal tissues in contact with it appeared to be undergoing degeneration.

“The site of the ovum appeared as a slightly raised convexity projecting into the uterine cavity, and at one point the epithelium was absent, there being here a mass of fibrin marking the spot where the ovum had apparently penetrated the maternal tissue.

“The trophoblast consisted of two kinds of cells — large mononuclear cells and multinuclear syncytia — and between these two types of cell all gradations could be traced.

“Beginning ingrowths into the trophoblast of the chorionic mesoblast could be seen, but no villi were yet formed. The embryo itself resembled that of the hedgehog at a corresponding period of development.”

Analogy with the ovum of insectivora and rodents, and the discovery of this unique specimen, together with the work of Selenka and Strahl, on placentation in the anthropoid apes, has established the fact that the human intra-uterine ovum imbeds itself by an active destruction of the mucous membrane of the uterus.

The necessity for obtaining pabulum for the growing ovum is thus met in man by means of destruction and absorption of the maternal tissues on which the ovum is engrafted, until a later period, when the formation of villi by ingrowths of vascular mesoderm into the trophic ectoderm enables the embryo to obtain nutriment by transudatory exchange.

This new view of the behaviour of the impregnated intra-uterine ovum (which has lately been strongly urged in this country by Teacher *) was soon applied to the ovum developing in the Fallopian tube, and the explanation of F1'irth’s seemingly anomalous case of tubal gestation was made clear by adopting it.

It must be remembered that the mucous membrane of the tube differs from that of the body of the uterus in having practically no submucosa, so that the developing ovum, if it behaves in the tube in a similar manner to that in the uterus, must imbed itself in the muscle on which the columnar epithelium of the tube practically stands.

Werth, Aschofi, Heinsius, Lockyer, William Duncan, and others have all brought forward evidence of the intra-muscular site of the tubal gestation, whilst Russell Andrews’ excellent paper on “The Anatomy of the Pregnant Tube ”[1] is responsible for the acceptance of the fact in this country that the early tubal ovum is entirely intramural in position.

But whilst the main fact that the human ovum is an imbedded one is now generally agreed upon by all competent observers, a number of lesser details are still under discussion, both as regards intra-uterine and extra-uterine pregnancy.

This is unavoidable, because the very early stages of the developing human ovum, either in the uterus or in the tube, have not been studied, even the specimen described by Peters being comparatively late from an embryological point of view, whilst in the tube the earliest specimen described is of at least 14 days’ growth.

The Authors' Speciments

Our own work on this subject is founded on the macroscopical and microscopical examination of eighteen specimens of early tubal gestation.

Amongst these we are fortunate to include two of 19, one of 24, and one of 30 days’ growth. Of these specimens one of the 19-day tubes and the 24-day tube have been cut in serial sections to the number of nearly 6,000. We are indebted to Mr. Murray for both these specimens. The 30-day tube was treated similarly. For the second 19-day tube we are indebted to Dr. W. Duncan, who has already reported the case, together with some pathological notes by one of us, founded on a large number of sections obtained from it. All these four pregnant tubes were specimens of “extra-tubal” rupture, and they were obtained as a result of the operations necessitated thereby.

One patient on admittance to the hospital was nearly dead, and an operation failing to save her life we were enabled to obtain the uterus and adnexa at the autopsy.

The two 19-day tubes each presented on the isthmic segment a small globular enlargement about 7mm. in diameter, in the antimesosalpingeal side of which was a perforation about 3 mm. in diameter with clean-cut edges. The ovum had escaped from the sac, but in neither case was found, although a careful examination was made of the blood removed from the peritoneal cavity.

The 24-day tube was similar in all respects except that the enlargement measured 10 mm. in diameter, and that it was situated in the ampullary segment. In neither of these three specimens was there any general distension of the tube.

The 30-day tube presented similar appearances. The gestation was in the ampullary segment. There was, in addition to the extratubal rupture, an intra-tubal rupture as well, and the lumen of the tube contained a blood-clot the size of an almond.

The Site of Tubal Gestation

The serial sections obtained from our specimens of pregnant tubes of 19 and 24 days’ duration entirely confirm the results published by Fiirth, Werth, Aschofi, Heinsius, Lockyer, and Russell Andrews, viz., that the early tubal getation is entirely intra-mural, and do not support the view of Couvelaire that the tubal ovum is never completely imbedded in the tube wall.

The serial sections obtained from our three early specimens of tubal gestation already referred to yield results so nearly identical that a detailed account of those obtained from one of them will suflice to demonstrate this point.

Berkeley1905 fig01.jpg

Fig. 1 illustrates a transverse section of the tube on the abdominal side of the gestation sac and presents nothing abnormal.

It will be seen from the simple arrangement of the plicee and the relative thickness of the muscle that it is taken from the isthmic portion of the tube. An attentive examination of the section reveals no trace of foetal cells, either in the tissues or in the vessels.

Berkeley1905 fig02.jpg

Fig. 2. illustrates a section taken some 0.7 mm nearer to the gestation sac. Fetal cells (trophoblast) are just beginning to appear in the inner portion of the longitudinal (outer) coat of the muscular wall of the tube.

These cells, which from their large size and deeply-stained nuclei strike the eye at once, are lying in and around somewhat crescentic spaces, formed by delaminations of the outer longitudinal muscle coat. The whole tube is thickened, this being partly due to increase in its muscular elements and partly to an appearance of oedematous swelling aflecting the connective tissue between the muscle bundles.

It will be noted that the earliest signs of the fcetal cells in this section are on the side towards the mesosalpinx.

Two large arteries cut in transverse section are seen where the tube joins the mesosalpinx.

Berkeley1905 fig03.jpg

Fig. 3 illustrates a section taken about a millimetre nearer the centre of the gestation. Evidence of foetal invasion is here very marked along crescentic delaminations of the muscle wall on the side of attachment to the mesosalpinx. Along these delamintations the troplioblast cells appear to be proceeding in such a way that the walls of the spaces are very deeply infiltrated by them. The maternal tisue in their vicinity is broken up, and is undergoing a fibrinous change. There is little or no evidence of proliferation of the maternal cells or of an active reply of any kind by them to the destructive action of the trophoblast, whose cell are mainly of two kinds—large mononuclear cells and irregular multinuclear syncytia. Here and there are seen multinuclear cells, which are probably to be regarded as intermediate stages between these.

Berkeley1905 fig04.jpg

Fig. 4 reproduces the appearances shown by a section taken a millimetre nearer the uterus.

A definite gestation sac is now seen, which lies to one side of the lumen of the tube, chiefly towards the attachment of the mesosalpinx. It is separated from the lumen by the circular coat of muscle.

The wall of the sac is formed by muscular tissue densely infiltrated by trophoblast cells (the “ cell-sheet ”). The sec surrounds the circular muscle coat for about one-half of its circumference, whilst around the remaining half the trophoblast cells can be traced efiecting a delamination of the muscle fibres, and particularly tending to split the longitudinal from the circular muscle coat. Wherever the maternal tissue is in contact with the cells of the trophoblast, evidence of extensive fibrinous degeneration is met with, the muscle bundles becoming indistinct and homogeneous, and eventually forming layers of somewhat concentrically laminated fibrin.

Note. — Figs. 1 to 9 were drawn, as regards their chief detail, with the Camera Lucida. These drawings were then enlarged with the Pantograph. The finer details were then filled in in a semi-diagrammatic manner. Finally, the completed drawings were reduced to their present size, the proportions between thembeing maintained.

In the process of invasion zones of muscular tissue appear to become sequestrated both from within and without by masses of trophoblast cells, and subsequently to undergo fibrinous change.

The gestation sac contains many chorionic villi, exhibiting the usual bilaminar epithelium, which in many places can be traced by oontinuity of cells into the layers of invading trophoblast.

The circular muscle coat appears to resist the fibrinising action of the trophoblast much more successfully than the somewhat loosely-structured longitudinal coat, so that the gestation sac is still separated from the lumen by a considerable thickness of unaltered circular muscle.

Numerous blood extravasations are seen amongst the more peripheral layers of the trophoblast cells.

The method of invasion of the tube wall by the trophoblast appears chiefly to be a process of delamination followed by fibrinisation, and in some places the walls of the maternal vessels are particularly singled out for attack, so that sections of vessels (usually veins) appear in which a large part of the circumference of the wall is replaced by trophoblast cells, the remainder appearing normal.

It is, however, often difficult to distinguish space formed by delamination of the muscle fibre from the lumen of veins. This much appears certain, viz., that the trophoblast cells do not grow along the veins, but rather tend to invade them from without, being never found occupying the lumen of blood-vessels, so that the peripheral spread of the trophoblast takes place by an intercalation of foetal cells between the muscular strata of the tube wall.

Berkeley1905 fig05.jpg

Fig. 5 represents a transverse section of the pregnant tube about 1 mm. nearer the uterus.

The gestation sac is now large, and is seen to be surrounding the circular muscle coat in a very remarkable way.

It is evident that the ovum has split the longitudinal coat from the circular coat and is lying between them.

The wall of the ac is composed of fibrinised muscle tissue deeply infiltrated with trophoblast cells.

Within the cavity lie many villi, covered by-the usual double layer of epithelium. At a few spots this can be seen to -be continuous with the layers of trophoblast cells, which constitute the “cell-sheet.”

No villi are found amongst the maternal tisues.

Large masses of syncytium are found here and there, chiefly in relation with the epithelium of the villi, with the inner cellular layer of which, (Langhans’ layer) they can be seen to be directly continuous.

Syncytia are always found in a space, either the gestation sac itself or some of the blood spaces which appear to adjoin it.

In many places masses can be seen made up of cells closely united, but yet exhibiting definite cell boundaries.

Probably these represent syncytia in which segmentation of the mass has taken place into definite though as yet unseparated cells. 84 Journal of Obstetrics and Gynwcology

At one point the circular muscle coat, with its contained tubal lumen, is joined to the separated longitudinal coat by a mass of fibrin with blood extravasations in its substance.

On the side of the tube, opposite the mesosalpinx, the gestation sac is bounded externally by a thin layer of the longitudinal muscle coat, which is extensively infiltrated by the trophoblast, and has undergone fibrinous degeneration throughout almost its whole thickness, the peritoneum alone remaining unaltered.

The lumen of the tube is still separated from the gestation sac by a considerable thickness of circular muscle. The tube itself is dilated and its plicze have disappeared. The columnar epithelium is more cubical in outline. Within it is seen a plug of fibrin of a circular shape, the periphery of which is made up of a number of small mononucleated cells. In this section the plug appears separated from the epithelium by a space around it. The circular muscle adjoining the gestation sac i fibrinised and infiltrated with trophoblast cells, though as yet to no great depth, except at one point where it has extended more deeply.

File:Berkeley1905 fig06.jpg

Fig. 6 Illustrates a section taken a little further up the tube. The general appearances are the same, but the rupture in the wall of the gestation sac is now seen at a point corresponding to the thinned and degenerate portion of the circular muscle noted in the preceding section.

The bulk of the gestation sac is still in the mesosalpingeal half of the tube, although the rupture is in the opposite half.

The lumen of the tube is still occupied by the plug of fibrin, covered externally by scattered mononuclear cells.

File:Berkeley1905 fig07.jpg

Fig. 7 is a transverse section taken through the centre of the rupture in the tube wall. This is seen to be still in the side of the tube opposite the mesosalpinx. The gestation sac is still bounded by layers of laminated fibrin, which can be seen being formed by the degeneration of the longitudinal muscle fibres, whilst partially imbedded in, but chiefly between them are dense masses of trophoblast cells. Many villi are present in the gestation sac, whilst a large mass of fibrin, into which blood collections have extravasated, is seen on the right hand of the drawing.

The lumen of the tube is cresccntic in shape owing to the projection into it of a dense mass of fibrinous mtaerial which is situated in the circular muscle on the mesosalpinx side of this coat. Blood extravasations are found in this mass of fibrin, which appears to be replacing the circular muscle at this point, and extends from the tubal lumen through the circular muscle to the “ cell-sheet” that forms the boundary of the gestation sac.

Intermediate sections between this and that last illustrated show that this mass of fibrin projects into the lumen for some considerable distance, at first being sessile on the wall, but later becoming entirely free, so that in Figs. 5 and 6 it appears as a detached mass in the centre of the lumen.

The cells that cover it, i.e., that intervene between the fibrin itself and the lumen of the tube, we have been at some difficulty in allotting to a definite histological place. They are certainly not epithelial in origin, the tubal epithelium being absent over that part of the tube wall. Nor are they fmtal cells, being much smaller than the trophoblast cells, and having fainter staining nuclei and less definite outline.

We believe them to be cells proliferated from the connective tissue of the tube wall — probably from the scattered sub-epithelial stroma cells noted by Andrews.

Fig. 1.—Transverse section of the pregnant tube between the gestational sac and the uterine cornu. .A. Mesosalpinx. B. Circular Muscle Coat. C. Longitudinal Muscle Coat.

Fig. 2.—Transverse section of the pregnant tube approaching the gestational sac A. Trophoblast cells invading the longitudinal muscle coat. Note the tendency to concentric delamination of the muscle fibres.

Fig. 3.—Transverse section of the pregnant tube. A. Mesosalpinx. B. Tropholxlastic infiltration. C. Muscle fibres undergoing a fibrinoid change. Spaces are appearing amongst the trophoblast. and syncytia are now seen.

Fig. 4.—Transverse section of the pregnant tube. A. Mesosalpinx. BB. Cells of the Trophoblast. C. Circular muscle coat. D. Longitudinal muscle coat. E. Villus. IF. Gestation sac. G. Muscle fibres undergoing fibrinoid change. H. Syncytium. The gestation sac, it will be noticed, is bounded by a. zone of trophoblast cells and librinined muscle constituting the “ cell sheet."

Fig. 5.—Transverse section of the pregnant tube. A. Mesoaalpinx. B. B. “Cell Sheet " C. Lumen of Tube diatended, and containing a mu: of fibrin. D. D. Gestation uc.

Fig. 6.—Transverse section of the pregnant tube. A. Mesosalpinx. B. Gestation sac. C. Mass of fibrin, containing blood extravasa tions. 1). The rupture in the sac wall. _ 19'. ‘‘Cell Sheet." /' ‘T

Fig. 7.—Transverse section of the pregnant tube. A. Mesosalpinx. B. “Cell Sheet." 6'. Gestation sac. D. Lumen of Tube. 1'}. Rupture. 1. Mass of fibrin, representing track of ovum. 0. Mass of fibrin, containing blood extravuations.

Fig. 8.—Transverse section of the pregnant tube. A. Villi. B. Gestation Snc and Rupture. C. Capsularis. D. Lumen of Tube.

Fig. 9.—Transverse section of the pregnant tube. A. Trophoblast Cells invading Longitudinal Muscle Coat. B. Lumen of Tube. This section being oblique, only a portion of the circumference of the tube is included in it.

Fig. 10.—A portion of the "cell-sheet" highly magnified. A. Cells of the Trophoblut. B. Gestation sac. C. Syncytium. D. Fibrin deposited from blood in intervillous space, and containing polynuclenr leucocytes. E. Muscle undergoing fibrinoid change. F. Cells of Trophoblast embryonising wall of maternal vein. G. Unaltered vein wall. H. Lumen of vein. Gestation sac

The Varieties of the Rupture of the Primary Gestation Sac.

Berkeley1905 fig11-16.jpg

Fig. 11.— Extra-tubal Rupture. Fig. 12.— Intra-tubal Rupture, with escape of the blood from the abdominal ostium. Fig. 13.— lntra-tubal Rupture, with retention of blood in the tube lumen (Hmmatosalpinx). Fig. 14.— Intra-mural Rupture. Gestation nc

Berkeley1905 fig18-22.jpg

Fig. 18.— Combined extra-tubal, intra-tulval, and intra-mural Rupture. Fig. 19.—Combined extra and intra-tubal Rupture. Fig. 20.—Coml)ined extra-tubal and intramural Rupture. Figs. 21 and 22. Combined intra-mural and intra-tubal Rupture. without and with retention of the blood in the tubal lumen.

If this be so, then these cells might be said to form an attempt at a “ Decidua Reflexa,” although the term is an objectionable one.

This mass of fibrin evidently marks the track along which the ovum burrowed to reach its intra-muscular site, and it is very interesting to remember that in the early intra-uterine ovum described by Peters a similar plug of fibrin was found at the point where presumably the ovum had burrowed into the submucosa.

Fig. 8, which represents a section taken lmm. nearer the uterine end of the tube, shows a similar condition in respect of the relation of the gestation sac to the lumen of the tube, that is to say, it is entirely outside it; but the circular muscle surrounding the tube lumen is now extensively destroyed on the side adjoining the gestation sac, which occupies not the entire circumference of the tube but its antimesosalpingeal half only.

But although the circular muscle is much thinned at this point, definite muscle fibres separate the gestation sac from the tube lumen.

From this point onwards, i.e., towards the uterine end of the tube, the sections show a diminishing amount of foetal tissue and a gradual restoration of the muscle fibre, until in Fig. 9 we see the structure of the tube almost completely normal, except for masses of invading trophoblast cells, which are insinuating themselves between the layers of the longitudinal muscle coat.

The epithelium covering the tube lumen is quite normal in appearance. Sections taken further on show a tube in every respect normal.

It will be noticed that neither foetus nor membranes are present, these having escaped on the rupture of the gestation sac.

The sections obtained from our second specimen of 19 days’ duration exhibit practically identical features, but since we were unable to cut the entire gestation in serial sections they can only be used to support the results obtained from the first specimen. In so far as they go they entirely concur with these.

The specimen of 24 days’ duration shows a more advanced stage in the destructive action of the trophoblast. The gestation had occurred in the ampullary portion of the tube, and the serial sections obtained from it agree with those previously described, in that at no point does the gestation sac communicate with the lumen of the tube.

The sections are more complicated and difiicult to decipher than those from the earlier tubes, but several points are specially well shown. The epithelium and plice of the tube have undergone no pathological change except at one part where, presumably, the ovum had made its way out of the lumen into the muscle tissue. Here is situated a dense mass of fibrin which bulges into the lumen, but in a less marked manner than that seen in the earlier tube illustrated in Figs. 5 to 7. There is no intra-lumenous plug of fibrin, and the tube is neither distended nor is its epithelium flattened, whilst the trcphoblastic infiltration of the muscle tissue is very much more extensive, there being but few portions where the invasion has not occurred. The structure of the “cell sheet” is very well shown. No trace of foetal rudiments or membranes are seen, these having, as in the previously described specimens, escaped when rupture occurred.

In our fourth specimen, which is of about one month’s duration, the gestation has occurred in the ampullary end of the tube close to its junction with the isthmus. The sac is situated in the wall of the tube and has ruptured into the peritoneum through a hole about four inches in circumference. Internally the sac communicates with the lumen of the tube by a large aperture, and in the lumen is a mass of blood-clot showing some chorionic villi on its surface. The gestation sac is empty, and its intramural site is very obvious to the naked eye in this specimen.

The remainder of our specimens of pregnant tubes consist of cases in which rupture of the gestation sac with blood extravasation has occurred, forming varieties of haamatosalpinx. In such specimen there is often difliculty in ascertaining the exact location of the gestation sac, as the tube is much disorganised by the blood extravasation; nevertheless, in those in which it has not been so destroyed by haemorrhage as to be unrecognisable, it is obvious that the gestation sac lies without the lumen of the tube.

Decidual Formation in Tubal Gestation

There has been much dispute in the past as to whether the pregnant tube forms a decidua in any way comparable to that of intra-uterine pregnancy. In the healthy tube there is practically no submucosa, the epithelium standing on the circular muscle coat directly. There is therefore, at the outset, a great difierence from the anatomical conditions obtaining in the uterus. Russell Andrews has described a sub-epithelial stroma in tubes the subject of chronic salpingitis, but these are the least likely to become. the seat of a pregnancy.

The term “decidua” can in tubal pregnancy be applied in three ways :—

1. In the sense of a definite massive proliferation of a (supposed) sub-epithelial stroma causing a decidual hypertrophy of the tubal mucous membrane analogous to that occurring in the uterine mucosa.

Such a condition is described by Orthmann, C. Webster, Whitridge, Williams, and others, and the two latter authors describe it in the opposite tube as well. In our specimens no such proliferation is present, unless it be represented by the scattered cells which surround the fibrinous plug.

2. In the sense of a “decidua capsularis” separating the ovum from the tube lumen, just as the decidua capsularis (reflexa) separates the intra-uterine ovum from the cavity of the uterus.

It is obvious that the muscle tissue intervening between the gestation sac and the tube lumen is in respect of the manner in which it has come to obtain that relation strictly comparable with the decidua reflexa of the uterus. There, however, the likeness ends.

In none of our specimens did this layer show any cell proliferation, but rather a tendency to degeneration, nor did it on account of the small size of the tube tend to bulge into the lumen.

Indeed, as can be seen by reference to the drawings, the gestation sac or implantation space in our earliest specimen extends right round the tube lumen, so that if we wished to define a decidua simply by homology of position with that in intra-uterine pregnancy it would be very difficult to know whether to describe the circular muscle separating the lumen from the gestation sac as reflexa or vera.

We think, therefore, that the term decidua should not be used in this sense.

3. In the sense of a proliferative reaction of the connective-tissue cells of the wall of the tube irrespective of their position.

The term “decidua” appears to be used by most modern writers in this sense.

Thus Russell Andrews describes isolated decidual cells in the folds between the plicse, and decidua-like cells scattered irregularly over other parts of the tube wall, but we agree with him that such cells do not merit the term “ decidua” when taken en masse.

In our sections the absence of connective-tissue reaction to the invading trophoblast is, perhaps, one of the most striking features. Here and there are patches of small cells with single rounded deeplystaining nuclei, whilst scattered irregularly about in the muscle tissue are certain cells with large oval vesicular-looking nuclei which stain faintly with haamatoxylin. These cells are certainly decidualike and are apparently also described by Teacher as occurring in the uterine wall, the seat of chorion-epithelioma, but with the exception of these the maternal tissues appear to be undergoing a passive destruction.

Decidual cells have been described by many authors as occurring in the ovary, the opposite tube, and the peritoneum, but we have been unable to find them.

Method of Imbedding of the Ovum in Tubal Gestation

In certain insectivores (Erinaceus) and myomorphio rodents the 88 Journal of Obstetncs and Gynaecology

ovum becomes sequestered in a crypt in the uterine mucosa, the mouth of which subsequently closes over it. The ovum subsequently grows at the expense of the maternal tissues surrounding it, which it destroys.

In the rabbit amongst lagomorphic rodents and the chiroptera the ovum appears to attain its sub-epithelial position purely by a process of burrowing, and from the appearances of the early human ovum described by Peters there was good reason to believe that burrowing had also occurred here.

In tubal gestation evidence as to the method by which the ovum first gets sequestered in the tube wall is not so clear, because sufficiently early specimens are wanting ; nevertheless in our specimens we think that the depth at which the gestation sac lies from the lumen of the tube, together with the fibrinous track indicating apparently the path of the ovum, strongly suggest that borrowing has taken place.

Of course the inter-plical crypts would ofier very favourable conditions for primary imbedding, in the same manner as obtains in the hedgehog, and the possibility of this cannot be denied; but even allowing it to have occurred, an assumption of the burrowing process is still necessary to account for the deeply intramural site of the gestation sac.

In our early specimens the bulk of the gestation sac lay in the mesosalpingeal half of the tube, and the fibrinous plug marking the track taken by the ovum also ran in this direction.

It is probable that the ovum travels towards that part of the tube where the nutrition (vascular supply) is most copious, and that in the majority of cases it is primarily implanted in the attached half of the tube wall.

Trophoblast Cells and Decimal Cells

The cells of the trophoblast are very characteristic objects, being of large size, irregularly polyhedral in form, with ob-ovate, or round, darkly,-staining nuclei. The cytoplasm also stains well. In the past there has been much confusion of these cells with decidua cells.

If one studies the characters of decidua cells, as seen in sections taken from the endometrium of the uterus in a case of tubal pregnancy, it is dificult to understand how confusion between these cells and those of the Trophoblast has come about. Decidua cells are considerably smaller, and are much more rounded in shape. Their nuclei are most markedly “vesicular” in appearance, and exhibit a fine, poorly-staining chromatin network. The cytoplasm stains poorly also, and between the individual cell bodies is a fine network which may receive processes from them.

If sections of placenta still attached to the uterine wall be made with a view to studying decidua. cells, it is seen that the cells graced by this name form a dense lamina interposed between the uterine muscle on the one hand and the chorionic villi on the other, and in sections such as these that we have made with a view to distinguishing between trophoblast and decidua. cells we find that this cell mass is attached to the uterine muscle by the medium of a scanty tissue of a loose fibroid character. The cells forming it present characteristics exactly resembling the cells of the trophoblast previously described, and bear no resemblance to the decidual cells found in the uterus of tubal gestation.

But there are other features strongly suggesting that this mass of cells is fatal and not maternal. It contains many layers of laminated fibrin, particularly on the aspect which bounds the intervillous space, and a reticulum of fibrinoid appearance lies between the individual cells, whilst direct continuity can be seen to occur between these cells and the epithelial cells which cover those villi which are attached to the cell mass (H a/tzotten or Fastening villi).

In short, the appearances of this mass of cells described by the older writers as the decidua serotina make it almost certain that it is identical with the cell-sheet bounding the early gestation sac. This has been emphasised by Kiihne, and we entirely corroborate him.

It is then no wonder that confusion should occur so long as these cells be termed “decidual cells” and their characteristics be made the criterion of decidual cells elsewhere.‘

The Method of Foetal Cell Invasion and the Structure of the Cell Sheet

As has been pointed out, the gestation sac (implantation space) is everywhere bounded by a layer of tissue composed of trophoblast cells and masses of fibrin.

The cells of the trophoblast present two main forms, the large mononuclear epithelioid cell and the multinuclear masses of protoplasm called syncytia.

‘Another very interesting point arises in this connection. In early abortion, either intra-uterine or tubal (cf. our sections), the entire cell-sheet remains in the maternal tissues after the expulsion of the embryo, membranes and villi, whereas the cell-sheet comes away with the rest of the fatal structures when the fully-formed placenta is expelled. Possibly it is owing to this fact that persistent hnmorrhage is commoner after abortion than after labour, quite apart from retention of gross products.

Of these the first is evidently the primitive one, for in addition to being much the most numerous it is the only one which can be said to invade the maternal tissue.‘

Sections taken from the periphery of the area. of foetal cell invasion show the mononuclear cell alone, syncytial masses only appearing after a definite cavity has begun to be formed; and the same may be said of the villi.

The method of invasion of these trophoblast cells has already been touched upon. They appear to be insinuating themselves between the muscle tissue of the tube wall in several strata, but always tending to a more or less concentric arrangement, whilst the layers of maternal tissue in contact with them appear to be undergoing a fibrinous degeneration. This fibrin, which is called after Nitabuch, who first drew attention to it, is further formed by a fibrinous material derived from the blood circulating in the gestation sac (intervillous space).

Since the trophoblast delaminates the musculariwall along many concentric planes, it follows that considerable areas of muscle tissue become segregated off between the layers of foetal cells and subsequently undergo fibrinous change. Thus is brought about the fact that the gestation sac is bounded by several layers of fibrin with foetal cells in between them, and this appearance is added to by the deposition of fibrin containing many polynuclear leucocytes from the maternal blood.

This admixture of fibrin and fcetal cells is known as the “ cellsheet’: (see Fig. 10), and in our specimens it is best marked in that of 24 days’ growth Here it is very deep, and the amount of fibrin contained within it is large. In our younger specimens it is not so well marked, and there is less fibrin.

Fibrin formation appears to reach a maximum in the first few weeks. Later, when villi form, it remains merely as a relic of the effects of the trophoblast cells, nutrition by destruction having given place to nutrition by transudatory exchange.

Stroganowa stated that Nitabuch’s fibrin was poorly formed in tubal gestation, but in our specimens it is found in large quantity. The same writer suggested that its presence in considerable quantity was a bar to the further invasion of maternal by fcetal tissue, and that the alleged absence of it in tubal gestation was the cause of the deep penetration and subsequent rupture of the tube wall. That this is not so is at once obvious, remembering that the cells of the “ cellsheet ” are already in large part outside the fibrin layer.

‘These observations are in accord. with the work of Hitschmann and Lindenthal, quoted by Russell Andrews.

Moreover, although the fibrinous deposit from the maternal blood may tend to strengthen the wall of the gestation sac, like the fibrin layers of an aneurysm, yet the part derived from the muscle tissue of the tube wall is an evidence of the destructive action of the fatal cells that, persisted in, must lead to rupture.

In all our three early specimens the splitting of the circular from the longitudinal muscle coat is very striking. No doubt this plane of potential cleavage presents a path of least resistance to the trophoblast cells. We shall show presently that it also may form a path of blood extravasation in that form of rupture of the gestation sac that we have termed “intra-mural.”

The Origin of the Syncytia

These multinuclear masses of protoplasm are always found on a free surface. Such free surfaces are in contact with maternal blood, either in the intervillous space or in the lumen of maternal vessels whose walls are being “embryonised” by the invading trophoblast. Thus syncytia may be said to depend upon blood contact for their production.

In some of the lower animals certain observers have described syncytia formation from maternal tissue (Selenka, etc.). We have failed to find any appearances indicating this, it being formed everywhere by a transformation of the mononuclear trophoblast cells (including Langhan’s cells), between which and the fully-formed syncytia all gradations can often be traced.

The “Embryonisation” of the Walls of the Maternal Vessels

Our sections show very clearly the appearances figured in Russell Andrews’ paper, viz., trophoblast cells replacing the tissues forming the walls of the maternal vessels (see Fig. 10). Nowhere does the foetal tissue grow along the lumen of the vessels in the manner described by some workers on placenta formation. It either directly invades the wall from without, so that the mass of trophoblast replacing the vessel wall is directly continuous with a similar mass lying outside the vessel, or vessels may be seen whose walls have been in part replaced by a cell mass not in continuity with any other part of the trophoblast. It is probable in these cases that the trophic ectoderm has found its way along perivascular lymphatics.

It is easy to see how readily the appearances presented by this “embryonisation” of the maternal vessel walls could be mistaken for an endothelial proliferation (trophospongia of Hubrecht).

Rupture of the Primary Gestation Sac

We have purposely used the expression “rupture of the primary gestation sac” in preference to “rupture of the tube,” because it has a much wider scope than the older term, including, as we shall presently show, three distinct conditions, whilst by the word “primary” we exclude rupture of secondary gestation sacs situated in the peritoneal cavity or broad ligament.

The Causes of Rupture

Rupture is brought about, excluding accidental mechanical causes, mainly in two ways. Of these the first is the destructive action of the trophoblast, which by the fibrinisation and absorption of the tube wall must of itself eventually lead to perforation.

The other factor is haamorrhage into the gestation sac from opened-up maternal vessels, which by causing a sudden rise in tension ruptures the sac wall already weakened by the destructive action of the trophoblast. This hsemorrhage is either into the intervillous space (which already contains blood) or into the space between the amnion and chorion, or into both spaces at once.

The second factor is by far the most common determining cause, as the study of any series of tubal gestations will show.

In our series only the three earliest failed to present unequivocal evidence of bleeding into the sac, whilst in these also there had most probably been some hsemorrhage into it. They were all specimens of uncomplicated intraperitoneal rupture of the sac, and in such cases evidence of pre-ruptural bleeding into the sac is almost impossible to obtain, because the blood escapes from the sac when rupture occurs.

The Directions of Rupture

Our specimens teach us that there are three directions in which the primary gestation sac may rupture. These three varieties of rupture are :—

  1. Extra-tubal rupture.
  2. Intra-tubal rupture.
  3. Intra-mural rupture.

They may be combined in several ways.

Extra-Tubal Rupture (Ruptured Tubal Gestation)

This includes all cases in which the gestation sac ruptures outside the tube (see Fig. 11). Berkeley and Bonney: Tubal Gestation 93

There are two sub-varieties. They are :—

(a) Intra-peritoneal rupture.
(b) Intra-ligamentous rupture.

Primary extra-tubal rupture is so well known as to need but a few words. It occurs most frequently in the isthmic segment of the tube for two reasons. Firstly,.that on account of the small size of the tube the trophoblast soon reaches the periphery; and secondly, because the lumen being relatively very small compared with the thickness of the wall, the gestation sac does not tend to bulge into it as is the case in the ampullary segment, and therefore intra-tubal rupture is uncommon here.

This variety of rupture is the one most likely to be brought about by the destructive action of the trophoblast alone.

We have suggested that the circular muscle coat ofiers a much greater resistance to the invasion of the trophoblast than the longitudinal coat, probably by reason of its greater density. In the isthmic segment of the tube this coat is relatively very thick. The growth of the trophoblast is therefore chiefly outwards, and it soon reaches the periphery of the tube.

Intra-Tubal Rupture (Tubal Abortion)

When the gestation sac ruptures into the lumen of the tube the condition is called tubal abortion (see Figs. 12 and 13).

This variety of primary rupture is much commoner in the ampullary segment of the tube, because the lumen being large here, the circular muscle coat thin, and the whole tube elastic and distensible, the gestation sac bulges into the canal.

Hemorrhage into the gestation sac appears to be the common determining cause of this variety of rupture, and the escaping blood either passes out of the tube into the peritoneum (haamoperitoneum, hamatocele, see Fig. 12), or is retained within the lumen of the tube (hmmatosalpinx, see Fig. 13).

The gestation may either remain in the sac or may be extruded along the lumen of the tube into the peritoneum, probably by blood pressure behind it, but possibly by active tubal contraction as well.

Intra-Mural Rupture

This form of rupture of the gestation sac has not, so far as we know, been previously described. It may be compared with the condition that obtains when a accular aneurysm becomes difiuse.

The gestation sac ruptures into the substance of the tube wall, and an extensive blood extravasation at once occurs, which forces its way amongst the muscle fibres until a large space is formed (see 14).

It will be remembered that the sections obtained from our early cases demonstrate very clearly the way in which the longitudinal muscle coat is split ofi from the circular muscle coat. Between these two coats there normally exists a potential plane of cleavage, and several of our specimens of later tubal gestation show equally clearly that in intra-mural rupture the blood extravasates along this plane of cleavage until it forms a mass which appears externally as a sausageshape enlargement of the tube.

These cases are almost invariably mistaken for hwmatosalpinges, the general outline and shape of which they resemble. A proper examination reveals the fact that the blood extravasation is separated from the lumen of the tube by a layer of muscle tissue more or less well marked.

In blood sacs of this nature clotting of the more peripheral portion of the blood is of common occurrence, so that the sac wall is somewhat similar to that of an aneurysm.

Rupture: of this blood sac frequently occur, and the contents then escape either into the peritoneal cavity, broad ligament, or lumen of the tube. The first of these is very common (see 15), and may occur at several spots, for the thin layer of muscle tissue and peritoneum, which separates the blood sac from the peritoneal cavity, soon perishes, and in many specimens the rough laminated deposit of fibrin, which is often found bounding the more fluid portion of the blood, alone intervenes between the latter and the peritoneal cavity.

Not infrequently these perforations are temporarily closed by fresh blood clots, a fact noted, though in a somewhat different light, by Stroganowa.

Rupture of the blood sac into the broad ligament, though much less common, probably occurs through none of our present specimens show it.

Rupture of the blood sac into the tubal lumen (see 16) is very common, the blood breaking through the circular muscle, and passing into the peritoneum via the abdominal ostium, or accumulating to form a true blood distension of the tube (haamatosalpinx).

Intra-mural rupture is much commoner in ampullary than in isthmic gestation, and it should be noted that the blood usually extravasates away from the uterus because of the looser structure of the wall of the ampullary segment.

In intra-mural rupture the ovum is not usually disturbed from its bed, and is to be looked for at the uterine end of the tubal enlargement.

It appears to us that some of the cases described as Para-tubal Hazmatocele by Sampson Handley[2] are possibly old examples of intra-mural rupture of a tubal gestation sac, and we were interested to note, whilst perusing this author's monograph, that Cullingworth described some years ago what was probably an example of intramural rupture, as an intra-mural haamorrhage due to a ruptured vein in the wall of the tube.

Combined Varieties of Rupture of the Gestational Sac

Combinations of the foregoing varieties of primary rupture of the gestation sac are very frequently met with.

This is explained by remembering that rupture is largely due to the tension in the gestation sac, and that if the occurrence of one variety of rupture fails to relieve the tension the gestation sac will probably rupture in some other direction. Thus in intra-tubal rupture (tubal abortion) the blood poured into the lumen of the tube is frequently prevented from escaping by closure of the abdominal ostium by impacted clot or inflammatory adhesion. The tension in the gestation sac not being relieved, extra-tubal rupture usually of the intra-peritoneal type may occur.

Some of the possible combinations of rupture are illustrated in Figs. 18 to 22.

A Note on the Aetiology of Tubal Gestation

This subject has recently been exhaustively reviewed by Russell Andrews. In so far as our specimens present any additional evidence bearing on the various theories which have been advanced, we may say that signs of pre-existing salpingitis are absolutely wanting in our early cases, which alone are of value in this connection.

Neither was there any suggestion of abnormalities, such as kinks, diverticula, etc., affecting the tubes which, both macroscopically and microscopically, were normal except at the site of the gestation. This was stated long ago by Bland-Sutton.

As regards evidence of “external wandering” of the ovum we would like to place on record that, in our three pregnant tubes of less than a month's duration, the corpus luteum was found in the ovary of the opposite side in two cases, and in the ovary of the same side in one case.

In conclusion, we would add that we have abstained from appending a complete list of references to the works of the authors quoted because such are to be found in the monographs of Russell Andrews, Teacher, and Arthur Robinson,[3] to all of whom we are indebted for material that is incorporated in this paper. Finally, we would express our thanks to Mr. A. G. R. Foulerton for the facilities he has afforded us during the prosecution of this research, and to Miss Appleyard for her valuable aid in the preparation of the drawings.

  1. Journ. of Obstet. and Gynaccol. of the British Empire, Vol. iv., p. 280.
  2. Journ. Obstet. and Gynacol. of the British Empire, Vol. i., p. 239. 96 Journal of Obstetrics and Gynaecology
  3. Hunterian Lectures, Royal College of Surgeons, England, 1903.

Cite this page: Hill, M.A. (2024, April 22) Embryology Paper - Tubal gestation - A pathological study. Retrieved from

What Links Here?
© Dr Mark Hill 2024, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G