Book - Contributions to Embryology Carnegie Institution No.56-8

From Embryology

Chapter 8. Hydatiform Degeneration in Uterine Pregnancy

Embryology - 13 Dec 2018    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)

Mall FP. and Meyer AW. Studies on abortuses: a survey of pathologic ova in the Carnegie Embryological Collection. (1921) Contrib. Embryol., Carnegie Inst. Wash. Publ. 275, 12: 1-364.

In this historic 1921 pathology paper, figures and plates of abnormal embryos are not suitable for young students.

1921 Carnegie Collection - Abnormal: Preface | 1 Collection origin | 2 Care and utilization | 3 Classification | 4 Pathologic analysis | 5 Size | 6 Sex incidence | 7 Localized anomalies | 8 Hydatiform uterine | 9 Hydatiform tubal | Chapter 10 Alleged superfetation | 11 Ovarian Pregnancy | 12 Lysis and resorption | 13 Postmortem intrauterine | 14 Hofbauer cells | 15 Villi | 16 Villous nodules | 17 Syphilitic changes | 18 Aspects | Bibliography | Figures | Contribution No.56 | Contributions Series | Embryology History

Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic Textbook" and "Historic Embryology" 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 and interpretations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)


To read even the titles of articles on "molar" pregnancies which have appeared during the last few decades is a rather wearisome task. The great majority of the articles concern themselves merely with the report of "a case" or (rarely) of "several cases" of hydatiform moles. The recent cancer literature stands in marked contrast to this, for not even the general practitioner would think of reporting a routine case of cancer of the breast, let us say. The significance of these facts is self-evident, and whatever else they may mean they do imply that hydatiform mole is still regarded quite generally as a rare condition. Indeed, many of those reporting "a case" frankly say so, and although the incidence of hydatiform degeneration is estimated variously by different authors and investigators, there seems to be entire agreement among them that it is a rare, even if not an extremely rare condition. This opinion seems to be shared also by those general practitioners whose long practice runs high up into the hundreds or even into the thousands of obstetrical cases. Indeed, many general practitioners declare that they have not seen a single case of hydatiform mole during the practice of a long life.


This prevailing opinion can not be attributed solely to the influence of the schools or to books, but is based upon the actual experience of the individual practitioner and upon his conception of what constitutes hydatiform degeneration. This is illustrated, for example, by Menu (1899), who said that a small hydatiform mole weighs 300 grams, a large one 8,000, with an average weight in his series of cases of 1,700 grams. But even specialists in charge of hospitals have reported experiences similar to that of the general practitioner. Pazzi (1909), for example, stated that although he had observed more than 6,000 cases of labor in his private and hospital practice, he never met with a case of hydatiform mole. Moreover, it would seem that only some specialists have come to regard the condition as somewhat less rare than was heretofore supposed. This is well expressed by Williams (1917), who wrote: "Hydatiform mole is a rare disease, occurring, according to Madam Boivin, once in 20,000 cases. On the other hand, the statistics of Williamson would indicate that it may be found but once in 2,400 cases." Williams adds, however, that in his own experience it occurred even more frequently than stated by Williamson; and Essen-Moller (1912), on the basis of 6,000 cases treated between 1899 and 1908, gives the incidence at the Frauenklinik at Lund as 3 per 1,000. My former colleague, De Lee (1915), in commenting on hydatiform degeneration, also stated that he "frequently found in aborted ova one or more villi degenerate and forming vesicles"; and similar remarks were made also by others, notably by Muller (1847), Marchand (1895), Veit (1899), van der Hoeven (1900), Hiess, and according to him also by von Hecker, Langhans, Weber, and Frank el. Findlay (1917) also regards "it as fair to conclude with Veit, Freund, and Dunger that abortive types of hydatiform mole are commonly overlooked," and although he gave no evidence for his opinion, Weber (1892) insisted that hydatiform mole "occurs much oftener than we are led to believe from books or other literature." Essen-Moller says Konig gave an incidence of 1 per 728 cases. Pazzi (1908 b ) stated that Dubisay and Jennin found in 1903 that hydatiform degeneration occurs once in 2,000 pregnancies, and that Cortiguera in 1906 declared that the frequency of hydatiform mole has a discouraging variation of from 1 in 3,000 to 1 in 700 labors, but that in his personal experience Cortiguera saw one case in every 300 labors. The latter incidence is only slightly higher than that given by Essen-Moller for the clinic at Lund, and somewhat below that of Kroemer (1907), who found 15 hydatiform moles in 3,856 births, or one in every 257 cases. Mayer (1911) reported 10 instances among 3,105 cases of labor, an incidence of 1 in 310 cases, and it is only necessary to add that Donskoj (1911) stated that the incidence of hydatiform mole in 28,406 cases at the Frauenklinik at Mlinchen, between the years 1884 and 1910, was only 1 for every 4,058 births, to emphasize the discouraging variation of which Cortiguera spoke. Donskoj also stated that Engel gave the incidence as 1 in 800, and Korn as 1 in 1,250 births. Such a surprising fluctuation in the apparent incidence in adjacent communities points to differences in conception of what constitutes a hydatiform mole, and also to differences in character of the material upon which the calculations are based.


The existence of hydatiform degeneration in far greater frequency than commonly supposed is indicated also by the records of the Department of Embryology of the Carnegie Institution of Washington. However, the material covered by these records is not identical with that upon which the above opinions or those of other obstetricians are based. The opinion of the obstetrician is based upon material belonging very largely in the later months of pregnancy, while that in the Carnegie Collection, on the other hand, belongs very largely in the earlier months. Hence this material is not truly representative of the entire period of gestation. But the same thing is true of the material upon which the general practitioner, the obstetrician, and the gynecologists have based their opinions, for these are based largely upon material from the last months of pregnancy. Hence, the cases of hydatiform degeneration which survive mainly come to their attention.


But unless we can assume that the incidence of hydatiform degeneration is constant during the whole period of gestation, its incidence at any particular time of this period may very incorrectly express that at any other time. This could fail to be true only if the incidence of death of the conceptuses and their susceptibility to hydatiform degeneration were exactly uniform throughout every period of intrauterine life. But we know that neither is true, for it is common knowledge that by far the great majority of the cases of uterine hydatiform degeneration recorded in the literature are mature specimens of total or partial degeneration obtained in the later months of pregnancy. Although such specimens may contain villi in various stages of degeneration, they nevertheless represent end or near-end results. Like the fetuses which rarely accompany them, they are full-term or near-term products when regarded as hydatiform degenerations, and unless we are to assume that conceptuses once affected by hydatiform degeneration always survive to near-term, statistical deductions based upon the cases that do survive can give us little idea of the actual frequency of the condition throughout the entire period of antenatal life.

That the specimens upon which past and also present opinion is based usually were large is confirmed by the belief in the prevailing clinical criterion of the existence of a disproportionately large uterus in cases of hydatiform mole. The emphasis laid on this by clinicians is well illustrated by Seitz (1904 b ), who said that cases in which the uterus is too small are the exception. Indeed, it seems that the validity of this clinical dictum has been questioned only very recently by Briggs (1912). Since most early conceptuses showing hydatiform degeneration have been inhibited in growth before being aborted, it probably is only the specimens which continue to grow that produce a uterine enlargement greater than could normally be expected. However, since, as emphasized by Gierse (1847), Storch (1878), Hiess (1914), and others, most hydatiform moles are expelled early and spontaneously, it is evident that these can not have been adherent that is, have penetrated very deeply or they would not have been expelled early and spontaneously. Furthermore, maceration changes so commonly present in aborted hydatiform moles indicate very clearly that a large percentage of them, together with the decidua, had been more or less completely detached from the uterine wall some time before abortion occurred.


As far as one can gather from the literature, the present opinion regarding the incidence of hydatiform degeneration would be paralleled quite correctly if, in the case of measles, we assumed that it was as common in octogenarians as in children. Measles, indeed, is an extremely rare disease in advanced age, but it nevertheless is very common in infancy. This is exactly the mistake we have made regarding hydatiform degeneration. It may be and undoubtedly is a rare disease at or near term, as Gierse also stated, but it probably is the commonest of all diseases during the earliest months of gestation. The typical large hydatiform mole is an endresult which it has taken long months to develop. No one seems to have followed its evolution, although hydatiform degeneration, whether total or partial, is, of course, gradual in its advent.


The records of the Carnegie Collection contained 8 cases of hydatiform mole in the first 2,400 accessions, showing a frequency eight times as great as that given by Williamson, or an excess of 700 per cent. Since the first 2,400 accessions contain 309 cases of tubal and 2 of ovarian pregnancy, only 2,089 uterine specimens remain. Hence the recorded incidence in the uterine specimens really is 8 in 2,089, or 1 in every 261 cases. This incidence is only slightly lower than that of Kroemer, and somewhat higher per 1,000 than that given by Essen-Moller for the Frauenklinik at Lund, or than the personal experience of Cortiguera.


The highest incidence of hydatiform degeneration previously reported is that of Storch, who estimated it as 50 per cent, but he unfortunately did not give a record of his cases. However, he emphasized that the typical, completely hydatiform mole is a relatively rare form of the disease, and that all manner of transition forms between the normal chorionic vesicle and the completely degenerated one can be shown to exist. Storch further emphasized the commonness of hydatiform degeneration, especially in the early months of pregnancy, but as Veit (1899) well said, Storch somehow has not received sufficient credit for his investigations and Gierse was forgotten completely. This seems strange, especially in view of the fact that Storch's work was done in Copenhagen, where Panum (1860) had done and still was doing such fine and very suggestive, indeed epochal, work on the origin of monsters. Although Storch devoted part of his paper to myxoma fibrosum, and reported only 5 cases of hydatiform mole, one of which, however, accompanied a living fetus, his opinions on the whole were far ahead of his time. In order to make this clear I shall quote a very significant passage, which, in the main, needs but slight changes to serve as a conclusion for my own investigations:

"Nun sind aber bekanntlich Eier mit blasiger Degeneration der Zotten und fehlerhaft oder nicht entwickeltem Fotus ein sehr haufiger Befund bei Aborten aus den ersten Schwangerschaftsmonaten. Mehrere solche Eier sind schon in den bekannten Arbeiten von Dohrn und Hegar beschrieben worden. Ich habe im Laufe des letzten Jahres eine grossere Anzahl von Aborten untersucht und derartige kranke Eier in mehr als der Halfte der Falle gefunden. Nicht selten ist die Amnionblase vollig leer und enthalt nur eine klare serose Fliissigkeit. In anderen Fallen sitzt an der einen oder anderen Stelle der Innenflache des Amnion ein kleiner rundlicher oder unregelmassig geformter, J/^-l Mm. grosser Korper, welcher aus Nichts als aus runden, schwach conturirten, zum Theil fettig entarteten Zellen und einer hellen, fast homogenen Zwischensubstanz besteht, und der durch einen feinen, 1-3 Mm. langen Strang von ahnlicher Natur mit dem Amnion verbunden ist. In noch anderen Eiern ist der Embryo zwar etwas weiter entwickelt, aber von den verschiedensten Formen von Missbildungen befallen. Seltener ist der Embryo einigermaassen wohl gebildet und von bis zu 2 Cm. Liinge, wie dies auch Hohl nur einmal gefunden hat. Sehr gewohnlich ist fettige oder lipoide Entartung des Embryo vorhanden; derselbe istdann eine kiirzere oder langere Zeit vor der Geburt abgestorben. Als die iiussersten Glieder dieser Reihe von kranken Eiern stehen endlich die sehr seltenen Falle, in welchen der Embryo seine Entwickelung ziemlich ungestort fortgesetzt zu haben scheint, und von denen die Falle von Martin und der oben beschriebene dreimonatliche abort Beispiele sind.
"Die blasige Entartung der Chorionzotten kann demnach neben den verschiedensten Zustanden des Embryo gefunden werden. Sehr hiiufig ist letzterer der Sitz von mehr oder weniger eingreifenden Krankheitsprozessen gewesen, die in demselben verschiedene Missbildungen hervorgerufen und ihn in seiner Entwickelung gehemmt haben. Es sind diese Krankheitsprozesse wahrscheinlich immer sehr friih im Ei entstanden, und miissen mit Panum zumichst als entziindliche Vorgange aufgefasst werden, welche nach ihrer Intensitat und vielleicht nach dem Zeitpunkte, zu welchem sie im Keime auftreten, bald eine theilweise Verodung der Keimanlagen der meisten wichtigeren Organe mit Verkriippelung des ganzen embryonalen Korpers, bald mehr locale Missbildungen einzelner Korpertheile hervorrufen konnen. Das Erstere ist in den hier besprochenen Aborten sehr haufig der Fall; der Embryo ist zu einem unformlichen Klumpen umgewandelt, dem die meisten Organe deren Keime durch Entztindung zerstort worden sind, giinzlich fehlen. Von diesen verkriippelten Amorphi finden sich in anderen Eiern alle Uebergangsformen zu mehr oder weniger entwickelten Missbildungen was auch Panum an einigen Beispielen nachgewiesen hat. Es Scheinen in der That die nicht zerstorten Keimzellen der verschiedenen Organe, nach dem ablaufe des Krankheitsprozesses, ihren urspriinglichen Entwickelungsplan mit einer oft merkwtirdigen Hartnackigkeit, so gut sie es konnen, festzuhalten. Von diesem Verhaltnisse liefern die bekannten herzlosen Amorphi, die durch einen Zwillingsbruder ernahrt werden und dadurch zu einer oft bedeutenden Grosse heranwachsen konnen, ein schlagendes Beispiel. In unseren Aborten sind zwar diese Amorphi, die keinen Zwillingsbruder zur Erhaltung ihres Kreislaufes gehabt haben, friihzeitig zu Grunde gegangen, und ihre Gewebsteile sind einer fettigen (lipoiden) Entartung anheimgef alien ; sie haben jedoch ihre Entwickelung eine Zeit lang fortgesetzt.
"Est is von den verschiedenen Verfassern vielfach von einer Auflosung der Embryo neu in der Amnionflussigkeit und von einer nachherigen Resorption derselben gesprochen worden. Ich glaube indessen, dass diesen Vorgangen eine sehr geringe Rolle beizulegen ist. Man findet in der That gewohnlich Nichts, was auf eine solche Resorption deuten konne. Es scheinen vielmehr die abgestorbenen Embryonen auch lange nach ihrem Tode eine grosse Wiederstandfahigkeit gegen die Einwirkung, von Amnionflussigkeit beizubehalten. Ich habe mehrmals ganz kleine, verkriippelte Embryonen zwar fetig entartet, in ihrer Form aber vollig wohl erhalten, in Eiern gefunden, die bis zu 10 Monaten im Uterus zuriickgehalten worden sind. Zudem ist die Amnionflussigkeit in diesen Eiern meist ganz klar, oder sie enthalt nur losgestossene, hinfallige Amnionepithelzellen suspendirt. Wenn daher die Eier ganz leer gefunden werden, so riihrt dies gewiss am Haufigsten daher, dass der Primitivestreifen seiner Zeit vollig destruirt worden und somit gar kein Embryo zur Entwickelung gekommen ist. . . . Im Allgemeinen erreichen sie kiene bedeutende Grosse und werden zudem oft friihzeitig aus dem Uterus ausgestossen, in dem sie, wie oben besprochen, ein sehr betrachtliches Contingent zu den Aborten iiberhaupt liefern. . . .
"Die Traubenmole und die verschiedenen Uebergangsformen derselben, die an Aborten sehr haufig vorgefunden werden, ist als Hyperplasie und secundiire cystoide Entartung des (von Allantois nicht herstammmenden) Chorionbindegewebs vorzugsweise charactertisirt. Die Krankheit wird von pathologischen Zustfinden der iibrigen Eitheile, Amnion und Embryo (Missbildungen, Verkriippelungen und friihzeitigem Absterben des letzteren) sehr haufig begleitet. Seltener ist der Embryo regelmassig entwickelt, stirbt aber meist auch dann wegen mangelhafter Vascularisation der (Chorion) Placenta friihzeitig ab. Sehr selten scheint der Embryo ungestort bis zur Geburt sich fortenwickelt zu haben."


But the unregarded observations and also the illustrations of Gierse, a few of which are reproduced in figures 94 and 95 (plate 7, Chap. VII), are still more startling than these opinions and observations by Storch, who knew of Gierse's observations published posthumously by Meckel. The latter quite correctly stated that such careful observations as those made by Gierse always introduce new points of view. If it be remembered that in these days, almost a century later, specimens of hydatiform degeneration which are 4 cm. in diameter still are reported separately as examples of early hydatiform degeneration, the great merit of Gierse's observations in this regard alone will be clearly evident upon recalling that he pictured a hydatiform villus from a chorionic vesicle the size of a hazelnut (about 12 mm.), the largest hydatids on which were only one-third of a line large. Moreover, Gierse added:

" Dergleichen geringe krankhafte Veranderungen finden sich an auserordentlichen vielen Abortus, und sie scheinen die haufigste Ursache des Abortus in den ersten Monaten zusein."


How such an epoch-making conclusion not only could be forgotten, but absolutely overlooked or disregarded, by all but a few of the scores upon scores who have written on hydatiform degeneration, it is difficult indeed to understand. Gierse, who also took steps to ascertain what normal villi look like, stated that villi with marked irregularities, as described by Desormaux, Breschet, Raspail, and Seiler, undoubtedly were abnormal, surmised that villi in abortuses seldom are normal, and added that between the slight pathologic changes in the caliber of the villi and the most evident hydatiform moles the plainest transitions can be found. Among other important things, Gierse also recognized the early fenestration of the stroma and pictured such a villus under a magnification of 250 diameters. Although reported very briefly, his findings, wholly confirmed here, still wait for general recognition.


Just as the great majority of specimens described in the literature are large, so 4 of the 8 specimens originally classed as such in the Carnegie Collection also are large, and none of the 8 is very young, as the following protocols show:


No. 70 (Dr. Charles H. Ellis) is a small, firm, degenerate-looking, almost solid mass 40 by 30 by 28 mm., composed of small cysts, degenerate deciclua, exudate, and degeneration products. It is very similar to a very much larger specimen, No. 323 (Dr. V. Van Williams), a large, firm, felt-like mass 120 by 90 by 65 mm. The individual cysts, which vary from 1 to 20 mm., are packed together rather firmly, though a few large ones are free. The exterior of the specimen is formed by a thick layer of degenerate decidua and gives only a slight indication of its true nature upon closer inspection or upon examination of the cut surface. No fetal remnants were noticed, and microscopic examination shows that the specimen is composed merely of a large hydatiform mass which was retained for a long time and then aborted in toto with the surrounding decidua and exudate.


No. 719 (Dr. G. C. McCormick), on the contrary, is a fresh, loose, typical hydatiform mass composed of loose hydatids of various sizes. As the specimen floats loosely in fluid, it fills a half-liter jar about two-thirds. A considerable portion of the hydatid cysts are glued into a solid mass by blood, exudate and decidua, which form a layer on the exterior.


No. 1323 (Dr. J. W. Schlieder) is also a large mass very like the preceding, and completely fills a liter jar. It is accompanied by much clot and composed mainly of a large, thickwalled, hemorrhagic, necrotic mass 80 by 50 by 45 mm., containing a large, thin-walled cavity 65 by 30 by 25 mm., which is broken at one end. This cavity, which apparently is that of the chorionic vesicle, is empty, smooth, and thinwalled, except where it is composed of a characteristic hydatiform mass.


No. 1325 (Dr. Fred R. Ford) is a small, irregular mass 40 by 33 by 20 mm., the exterior of most of which is formed by a thin layer of decidua. Within this is a small group of quite typical hydatid cysts, the largest of which measures about 10 by 5 mm. The appearance of the specimen suggests that it is merely a fragment, though the amount of decidua present indicates that the entire specimen probably was not much larger. The history of this specimen is especially interesting because of the diagnosis of tubal pregnancy, caused by the presence of a cornual myoma and the occurrence of repeated bleeding.


By far the most interesting specimen, in some respects, of hydatiform degeneration among those diagnosed as such upon gross examination, is No. 1640. This abortus, received through the courtesy of Dr. J. W. Williams, measured 40 by 20 by 15 mm. Upon examination, Dr. G. L. Streeter found it to be composed of a flattened decidual and chorionic mass which, upon section, showed "pearl-like vesicular enlargements which suggest hydatiform degeneration." The exterior of this specimen is composed of a thin, hemorrhagic decidua which completely surrounds the villi. The hydatid nature of this clearly is recognizable upon close scrutiny with the unaided eye, and easily becomes evident upon magnification of 12 diameters with the binocular microscope. Examination of the histologic preparations reveals it to be a very fine specimen of relatively early hydatiform degeneration.


No. 1914 (Dr. G. C. McCormick) is a fine, very characteristic mass. It is like Nos. 749 and 1323, but very much larger, for in fluid it completely fills a 2-liter jar. This specimen was said to have accompanied a living 7-months' fetus, having been expelled between the fetus and the placenta. Only a small amount of clot, and what seems to be a small portion of placenta and membranes, accompanied it. Since the placenta was not saved, it is impossible to say whether the mass resulted from partial degeneration of the placenta belonging to the living child or whether it represented a degenerate twin placenta, which is rather unlikely but not impossible, in view of the well-authenticated cases found in the literature. This specimen is of interest not only for the numerous large, clear cysts, one of which measures 30 by 25 mm., which it contains, but because it accompanied the birth of a living child and because of the relative rareness of such a coincidence. In regard to the latter, Dr. McCormick added that in his experience of over 1,000 labors he had never before met this coincidence. The rareness of the specimen is emphasized still further by the statement of Professor Williams that such an instance has not been observed in a series of over 17,930 osbtetrical cases from the Department of Obstetrics of the Johns Hopkins Medical School, as well as by the small series of such cases recorded in the literature.


No. 1926, a companion specimen to No. 1640, is composed of material from curettage received through the courtesy of Dr. Karl Wilson, of the Department of Obstetrics of the Johns Hopkins Medical School. It was removed from the same patient about a year after specimen No. 1640. Upon gross examination the hydropic nature of some of the villi is plainly evident, and upon microscopic examination the diagonosis of hydatiform degeneration could be confirmed, although the villi were extremely degenerate. The menstrual history of this case fortunately is known and is thoroughly reliable. The last menstruation occurred January 24 and curettage was done August 4. Bleeding occurred every two or three weeks during March and April and was repeated throughout May. Since the uterus, which had reached the symphysis, had not enlarged any for months, in view of the long duration of pregnancy the operation was performed. The major portion of the specimen is very small. The chorio-decidual portion was felt-like in consistency and extremely fibrous,


This short review of the gross appearance of the cases of hydatiform degeneration recognized by the unaided eye with the customary criteria, originally classed as such in the Carnegie Collection, shows that they vary decidedly in their gross, naked-eye characteristics, both as to size and appearance. No. 1640 scarcely is distinguishable as a case of hydatifo m degeneration from gross appearances alone, unless one's attention is directed especially to the matter, but all the rest of the specimens, both small and large, not only are easily recognizable, but are so characteristic that they could not possibly be overlooked. As was indicated above, the incidence of these specimens of hydatiform degeneration among the first 2,400 accessions in the Carnegie Collection was 1 in every 261 abortuses, or more than 8 times the incidence given by Williamson, and 1.3 times that given by Essen-Moller. Although this incidence is so much higher, it does not necessarily contradict the statements of Williamson, for it represents the incidence of hydatiform degeneration in abortuses belonging very largely below 7 months, it tell the whole story for these months, for since the incidence of hydatiform


due largely no doubt to the long retention. Most of the accompanying material looks like mucosa rather than decidua, although some of the larger pieces very evidently contained villi. Some of these were relatively thick and fibrous, and others were vesicular. All of the material was extremely fibrous, making it difficult to get a satisfactory teased preparation. Accompanying this material was a small body 5 by 7.5 by 30 mm. Both nodule and stalk contained some remnants of the embryo. Although the appearance of the stalk suggests the umbilical cord, it contains fragments of the body of the embryo, some of which evidently are composed of nerve tissue.


Microscopic examination of the pedunculated mass further shows it to be composed of degenerate remnants of organs, tissues, and cells. It is partly denuded and partly covered by a layer of fibrous connective tissue which contains local thickenings. In other areas this fibrous layer gives place to a single or more-celled layer, or to polygonal, epithelioid cells. The interior of this specimen is composed of a degenerate jumble including fragments of the central nervous system, the heart, liver, and cartilages. The entire body is chaotic in its structure, and small fragments of the nervous system are scattered throughout its entire extent. This would seem to indicate that the disruption of the tissues was mechanical. The material in which the organic remnants are contained is composed of coagulum, some mesenchyme, cellular detritus, blood, and polymorphonuclear leucocytes, degenerated cells, which appear to have been phagocytic, but which are more likely fusion products or "symplasma" (as Bonnet called them). A few remnants of vessels are found only in the fragments of cartilage.


degeneration given in the records of the Carnegie Collection is based upon determinations made essentially in the usual way that is, by unaided inspection of the gross specimen alone we must regard it also merely as an apparent, not as the actual, incidence. For, as will appear later, the actual incidence can be revealed only by a careful gross and microscopic study of all specimens, both normal and pathologic. Such a study has not as yet been completed, but 348 uterine specimens classed as pathologic, and 108 pathologic tubal specimens, contained in the first 1,200 accessions, were carefully examined.


The actual number of cases of hydatiform degeneration found among the 333 chorionic vesicles of uterine abortuses classed as pathologic in the first 1,200 accessions was 105, or 31.5 per cent of the whole. This includes some doubtful cases, but a revision probably would add more than it would exclude. The incidence of hydatiform degeneration in the pathologic tubal pregnancies was somewhat higher even, or 45 specimens of undoubted hydatiform degeneration out of 108, or 41.7 per cent. Since nearly all the tubal specimens are young, while the uterine series contains many more relatively older ones, the effect of this fact upon the determined relative incidence of hydatiform degeneration among the pathologic tubal and uterine specimens must be borne in mind. For a reliable conclusion regarding the relative incidence in the uterine and tubal pregnancies it would be necessary to select a series from each, composed of specimens of approximately corresponding ages. What the incidence of hydatiform degeneration is among the uterine and tubal specimens classed as normal I do not know, but it undoubtedly is far below that in those classed as pathologic. It is well to remember, however, that many, if not most, of the instances of beginning degeneration very likely will be found among the specimens classed as normal. This is well illustrated by a hysterectomy specimen, No. 836, represented in figures 96 and 97.


If we assume that the incidence of hydatiform degeneration among the pathologic specimens in the rest of the Carnegie Collection is the same as that among those in the first 1,200 accessions, then we get over 375 estimated instances of hydatiform degeneration in pathologic tubal and uterine cases alone. Since I have found a number of chorionic vesicles accompanying embryos classed as normal which also show hydatiform degeneration, this number would be increased still further; but unfortunately too few of the specimens classed as normal were examined to justify an estimate. Yet these normal specimens form 60.4 per cent of the first 1,000 and 40.7 per cent of the first 2,500 accessions. This supposed increase, due to inclusion of specimens contained among the normal, would be offset somewhat, however, by the fact that the first 1,000 accessions contain a somewhat larger proportion of young conceptuses, each succeeding 1,000 probably becoming somewhat more representative of actual life conditions. The difference between the composition of the first 1,000 accessions and that of the 1,000 between 1,500 and 2,500 is not very great, however, for the former contains only an excess of 17.6 per cent of cases falling in the first five groups of Mall's classification, which groups are composed largely of specimens below an embryonic length of 20 mm. Then, the relative proportions of tubal and uterine specimens in the different thousands also must be taken into consideration. But in any case, the estimated incidence of hydatiform degeneration in the Carnegie Collection, calculated without regard to those contained among specimens classed as normal, is 8.7 per cent, and the actual incidence hence is probably more than 1 in every 10 accessions.


Estimated on the basis of hydatiform degenerations found among the uterine and tubal specimens classed as pathologic alone, the incidence would be 12.5 per cent of the first 3,000 accessions. Upon this basis the incidence among the uterine specimens alone would be 9.9 per cent, and among the tubal alone 28.1 per cent. This difference of 200 per cent between the tubal and uterine specimens may have some significance in connection with the cause of hydatiform degeneration. That this estimate is not altogether too high is indicated also by the last 392 accessions, the first gross examination of which, made by others than myself, revealed 21 hydatiform degenerations, or an incidence of 5.3 per cent. Since these accessions contained a disproportionately large number of older, naked fetuses unaccompanied by secundines, aborted during influenza, this incidence of hydatiform degeneration undoubtedly is too low. Moreover, no histologic examination has as yet been made. Nevertheless, the incidence among those accompanied by secundines and classed as pathologic was about 14 per cent and among those classed as normal about 4 per cent.


If, as alleged by various investigators, the great majority of abortions occur in the first three months, it is highly probable that many of these early conceptuses are lost and never come to the attention of anyone, and that therefore the proportion of early specimens in this or any other collection is no doubt too small. Moreover, in quite a number of specimens of the first 1,000 accessions the chorionic vesicles were too degenerate for examination and in others they were absent, but we have reason to believe that this is not true to the same extent in the material beyond the first 1,000 accessions. Then, too, since only a few relatively large sections from a single portion of the chorionic vesicles were examined, it is evident that some cases in which the degeneration may have been purely local were probably overlooked. Hence the actual as compared with the recorded incidence of hydatiform degeneration in this collection is probably not merely 8 times but 240 times as great as that given by Williamson (1900), and 33.3 times as great as that given by Essen-Moller.


Most persons will, I presume, be willing to regard an increase of 700 per cent above that of Williamson as possible, but one of 24,000 per cent above Williamson, or even 3,333 per cent above that of Essen-Moller, as wholly out of the question. Yet, strange as it may seem at first sight, this is not a random guess, but an estimate based upn the actual incidence of hydatiform degeneration as determined by a careful gross and microscopic examination of mounted and unmounted material from over 400 abortuses. However, I lay no special emphasis on these percentages, and am using them merely to emphasize the great frequency of hydatiform degeneration. It matters little whether we shall ultimately determine an incidence of 10 or 5 per cent, but it does matter considerably whether we regard the frequency as 5 or 0.05 per cent, for this is a difference in frequency of 10,000 per cent.

In view of the prevailing opinion, I realize that these findings may seem incomprehensible and perhaps incredible, unless it is distinctly borne in mind that it is not stated that this incidence refers to the later months of pregnancy or to term. What the incidence in the later months of pregnancy may be I do not know, but I have called attention to an apparently well-founded belief that it is a relatively rare condition, the estimates ranging from 1 in 2,000 to 1 in 728 or 300 cases.

In regard to the incidence of hydatiform degeneration in uterine specimens, it should also be remembered that the life, in contrast to the laboratory, incidence for the entire period of gestation is higher, not only because the chorionic vesicles were not included in many of the accessions and because others were too degenerate, but because I have not as yet been able to recognize the very earliest stages with entire certainty. Furthermore, many instances of hydatiform degeneration from the early months of pregnancy, especially the first and second, are inevitably lost. The increase due to these things would be offset somewhat, however, by the lower incidence of hydatiform degeneration in specimens from the last months of pregnancy, relatively few abortuses from these months being contained in the Carnegie Collection.


To what extent the material in this collection is truly representative of actual life conditions is difficult, if not impossible, to determine. This question could be answered only if all the abortuses and material from abortions actually reached physicians, and if the latter sent all of them to the laboratory. My own impression so far is that the material representative of a sufficiently large community probably would have a somewhat lower incidence, notwithstanding the fact that many specimens, not only of hydatiform degeneration, but of abortuses in general, especially from the first month of pregnancy, are lost. However, since the presence of hydatiform degeneration is especially common among early specimens, the inclusion of these might raise the incidence for the whole period of gestation more than the inclusion of all specimens (not excepting those of the last 3 months) would lower it. But the result obtained would represent the incidence of hydatiform degeneration in abortuses alone, and not that in all pregnancies. The latter could be obtained only by including all gestations which end normally. If we accept Pearson's (1897) estimate that approximately 40 per cent of all pregnancies end prematurely, then the incidence of hydatiform degeneration among abortuses would represent very nearly twice that in all pregnancies. Mall's (1908, 1918) estimate of 20 per cent prenatal mortality, on the other hand, would give us an incidence only one-fifth as great as that among abortuses. Hence, the actual life incidence of hydatiform degeneration in all gestations would then be 1 in 10, as based upon Pearson's, and 1 in 25, as based upon Mall's estimated prenatal mortality. But even if, as estimated upon this basis, 4 or 10 per cent of all conceptions end in hydatiform degeneration, this does not necessarily contradict the current opinion regarding its rareness at or near term.


A careful examination with the binocular microscope of all specimens has shown that hydatiform degeneration as a rule is sufficiently general even in young vesicles, so that sections of a single portion about 10 mm. square enable one to make a fairly reliable diagnosis. Now and then, however, the process seems to be rather irregularly developed, especially in the larger specimens.


In order to determine accurately the question of distribution of hydatiform degeneration over various portions of the chorionic vesicle, it is necessary to examine a series of sections of portions of the chorionic vesicle for each small specimen. This has not yet been done, but since the portions used for microscopic examination had been taken at random without previous knowledge of the existence of hydatiform degeneration in any but the 8 specimens so recorded and above described, and since over 450 vesicles were examined, I can not believe that it can often be limited to any particular small area on relatively young vesicles. In these it usually is quite general even if not complete. It is of special interest in this connection that Muggia (1915), after reviewing the small list of cases of alleged hydatiform degeneration of the chorion Iseve in connection with a study of a case of his own, came to the conclusion that these cases are not really degenerations of the chorion laeve, but merely partial degenerations of the placenta. Although I have given no thorough attention to the normal changes in the chorion Iseve, I am quite certain that they are not the cause of confusion in the series of hydatiform degenerations from the Carnegie Collection. Cases in which whole chorionic vesicles exquisitely hydatiform in character were contained in the tubes, and a number of others which still were implanted within the uteri showed equally exquisite hydatiform changes around the whole perimeter. Such cases as these ultimately confirm the opinion that in young vesicles the condition is, as a rule, general, except perhaps at its very inception. This is true particularly by the time the degeneration has reached a stage which can be considered at all typical in its gross development, as determined by careful examination of numerous specimens with the binocular.


It is especially interesting that, just as soon as typical, hydatid, elliptical villi or portions of the same begin to appear, the condition can be recognized with some certainty under a magnification of 12 to 20 diameters with the binocular microscope. It often was surprising how relatively early stages could thus be detected and the diagnosis confirmed later by histologic examination. Indeed, celloidin blocks of tissue from which sections had been cut gave splendid testimony when examined in fluid with the binocular. One of the not very early stages contained in utero and represented in figure 98 could be recognized with the unaided eye; and when examined with the binocular, under a magnification of about 12 diameters, the picture was unusually fine and wholly unmistakable, as shown in figure 99.


That hydatiform degeneration is incomparably more common in the earlier than in the later months of pregnancy, thus justifying the comparison made with measles, is substantiated by statistics covering the material examined. From these it is evident that, excepting cases of large hydatiform masses originally classed as hydatiform degeneration from inspection of the gross specimens alone, practically all of the specimens are relatively young. This is true especially of those from tubal pregnancies, and we may hence regard it as established that hydatiform degeneration is a change which is exceedingly common in the earlier months of pregnancy, just as measles is common in childhood, and that it becomes progressively less common as the end of pregnancy is approached, just as does measles as senility is approached. The obstetrician does not see most of the cases of hydatiform degeneration, for they are merely reported as miscarriages and the specimens often are destroyed or retained unrecognized by the general practitioner or the midwife. They are often aborted spontaneously and completely with the decidua and rarely are still contained in a closed decidual cast when they reach the laboratory.


The spontaneity of the abortion, especially in early cases, was emphasized also by Storch (see page 208). Cortiguera (1906) is reported by Pazzi (1908 b ) also to have declared that many moles disappear wholly without leaving a remnant, even if occurring repeatedly in the same woman, and Donskoj (1911) also stated that many of those aborted do not come to the attention of physicians because of their harmlessness. This, however, does not imply that those which persist and develop into large masses are equally harmless, and it must be remembered that it is upon these that the current opinion regarding the tendencies to malignancy of the hydatiform mole is based.


The conclusion regarding the greater incidence of hydatiform degeneration in the early months of pregnancy is conclusively confirmed by its occurrence in 33 of the 45 tubal specimens, within the first two classes of the pathologic division of Mall, and in 100 of the 105 uterine specimens in the first six classes of this division. Most of the specimens in these classes are composed of villi, of empty chorionic vesicles, or of vesicles with embryos most of which have a length of less than 20 to 30 mm. That hydatiform degeneration is more common in the early months of pregnancy is indicated also by the well-known reports of Kehrer (1894) on 50 cases, and of Borland and Gerson (1896), who found that 63 per cent of 100 cases had aborted in the fourth and fifth months of pregnancy. According to Seitz, Hirtzman (1874) also found that 62.8 per cent of 35 cases had aborted between the third and six months. Only 2 of Kehrer's 50 cases, or 4 per cent, and only 3 per cent of the cases of Borland and Gerson aborted at the tenth month. Bonskoj stated that 7 of the 10 cases reported by him aborted in the fourth month and none after the sixth month. He stated further that 56 per cent of Bloch's 50 cases aborted before the sixth month, 44 per cent later than this, one being retained until the fourteenth month. The latter case is especially interesting, because retention not only beyond term but after the death of the mole, seems to be regarded as relatively rare. This, however, does not imply that retention beyond the period of growth of the hydatid mole does not occur, although Sternberg (1910), who also emphasized the great rarity of this condition, erroneously stated that the German literature reveals only a single instance of missed abortion in case of hydatiform mole, viz, that of Poten (1901). In this case a hydatiform mole of the size of a duck egg was said to have been aborted approximately one month beyond term. Hence growth must have ceased long before and the mole have remained in utero as a "harmless body." To this case of Poten, Sternberg added a case in which a hydatiform mole 14 by 9.6 by 4.3 mm. was aborted in the twelfth month after the cessation of menstruation. Although Sternberg included 4 cases from other countries among these missed-abortion moles, viz, those o Sheil (undated), Ferguson (also undated), Colorni (1908), and Gaifani (1908), one can hardly doubt that more cases could be added. Since the case of Sheil was one of twin pregnancy, in which one conceptus became hydatiform, it is njt at all unlikely that some other cases among this rather small series of twin pregnancies accompanied by hydatiform degeneration may belong in this category.


Mayer (1911) also emphasized the fact that, although instances of retention of fetuses are very common, instances of retention of hydatiform mole are very rare, only a few cases having been recorded. Maj r er referred to 2 cases by Kehrer, 3 of Borland and Gerson, and 1 case of Lange, and reported 4 of his own. These 4 were found among 10 cases of hydatiform mole, an incidence of retention of 40 per cent. They are interesting, especially in connection with the observation of Briggs (1912) that, contrary to current belief, uterine enlargement often is not beyond the normal. Mayer reported that this enlargement was too great in but 1 of the 4 cases, and that retention lasted as long as 4 to 5 months.


At least 3 of the cases of hydatiform mole among those originally recorded as such in the Carnegie Collection belong among retained specimens, as the illustrations alone suggest. But a fair percentage of detached chorionic vesicles included in the list of cases here reported undoubtedly also was retained after the cessation of growth, and it is for this reason that I further emphasize the fact that the uterine volume in a considerable percentage of these cases also, instead of having been too great for the duration of the pregnancy, unquestionably was too small. This is well illustrated by the histories of specimens Nos. 70, 323, 1640, and 1926, and by the specimens themselves.


The average menstrual age of 51 of 112 uterine specimens of hydatiform degeneration in which the data were available was 66.6 days, or 2% months. As will be seen, this is a far lower average age than heretofore reported, a difference which explains itself from what has already been said. It is interesting that the average menstrual age of 5 of the 8 specimens in the Carnegie Collection originally classed as hydatiform degenerations is 168.2 days, or 1}/2 times as great, thus being in substantial agreement with the usual results. Three of these 5 are large specimens, the fourth measures 40 by 20 by 15 mm., and the other is composed of small fragments contained in material from curettage. From this alone it follows that the menstrual age is a very uncertain guide, especially to the size of a hydatiform mole.


It may seem superfluous to add anything to the good descriptions of the gross appearance of the typical hydatiform mole found in the literature. Such cases are so characteristic that even a novice can recognize them at sight. Yet if the findings reported here are reliable, or even approximately so, it nevertheless must be evident that, in the past, the great majority of specimens of true hydatiform mole have remained unrecognized merely because they did not happen to present the customary, well-known picture to the unaided eye. Small chorionic vesicles, such as No. 2077 (shown in natural size in figure 100), which attract no attention upon cursory inspection, may, and often do, present the most exquisite picture of hydatiform degeneration when seen under a magnification of 3 to 20 diameters, as illustrated in figure 101. This is true especially if the examination is made with the binocular microscope. Since I have adopted this method of examination it has been possible to recognize instances of decidedly general and typical hydatiform degeneration in chorionic vesicles less than 2 cm. in size, with later confirmation of the diagnosis by a histologic examination. However, I have not been able to recognize very early stages merely by examination of the gross specimens, for gross recognition is possible only when portions of at least some of the villi have become sufficiently elliptical or globular to attract attention. Histologic recognition is possible far earlier than this, as shown in figure 102.


The general appearance of the whole chorionic vesicle is sometimes an aid in gross identification, for the villi not infrequently are smooth, slightly branched, and unusually long, so that the vesicle looks shaggy, as illustrated in figure 103. Several hydatiform villi from this specimen are shown in figure 104. The typical gross hydatid or watery, translucent nature of the villi can not be relied upon in early stages, for normally shaped villi that have undergone considerable lysis may be almost transparent and also somewhat more than normally bulbous. However, save in the case of some specimens of tubal pregnancy, the swelling of the villi, due to maceration or to luetic changes, is quite different in character from that characteristic of hydatiform degeneration, and usually quite easily distinguishable from it. Judging from several specimens of villi which were macerated in distilled water during a period of weeks, post-partum maceration never should cause confusion, and the same thing undoubtedly is true of intrauterine maceration. The differences in appearance between macerating villi with disintegrating epithelium, stroma and blood-vessels, and others undergoing hydatiform degeneration, is well illustrated by villi from No. 640, shown in figures 105 and 106. Though photographs can not register all the distinctions, the contrast is so marked in this case that one can not fail to notice it.


Since numerous trophoblastic nodules are present also in other conditions, notably in retained placentae, as found by Aschoff and others, I have not been able to regard their presence in unusual numbers, in some cases of hydatiform degeneration, as of crucial value, but the absence of placental differentiation at a time when it should be present, with a uniform and unusual development of the villi over the whole exterior of relatively large chorionic vesicles, is decidedly significant and has often been found to imply the presence of hydatiform degeneration. The same thing is true of a very irregular distribution of the villi, or of uniformly distributed fusiform enlargements on the villi and of the loss of the dull appearance of their cut surfaces, as seen under the binocular. As soon as the stroma becomes hydatiform, and even before liquefaction is present, the cut surfaces of hydatiform villi look somewhat shiny and waxy or, perhaps better still, paraffine-like. This was well shown in a previous paper (Meyer, 1920, fig. 21). A bluish tinge is always present, and this appearance is very characteristic. However, how easily a specimen of hydatiform mole can be recognized by examination with the binocular alone necessarily will depend also upon the condition of the specimen. If the villi are matted, glued, or macerated, not only the early hydatiform changes, but even fairly advanced ones, are often masked so completely that recognition is difficult or impossible without histologic examination.


In many early specimens the diagnosis could be made at sight from a histologic preparation under low magnification, even when it was impossible to make a diagnosis by examination with the binocular microscope alone. The field of sections of the villi also looks more scattered and the caliber of the villi shows greater variations. What further makes this possible is not, as has been generally assumed since Marchand's epochal work on chorio-epithelioma, the appearance of the syncytium or that of the Langhans layer or of the trophoblast, but the changes in the stroma, which precede those in the epithelium. The evidence in regard to this matter is overwhelming, and in the early stages, when the stroma already has been altered, it often is impossible to tell whether the epithelial development is normally or abnormally active. Moreover, in spite of Marchand's conclusion, extremely large hydatid vesicles often have but a single smooth layer of epithelium. This has been asserted repeatedly by other investigators also. The two layers of epithelium are not by any means always present, and, while there is no agreement in the matter, the opinion nevertheless seems to be that the grade of epithelial proliferation can not be used as a criterion for the determination of the presence of hydatiform degeneration.


Langhans (1902) also stated that Marchand overemphasized the presence of epithelial proliferation, and rightly declared that all sorts of gradations occur between normal and hydatiform villi. Indeed, unless hydatiform villi invariably arise as such when the earliest villi appear, or arise de novo later, all gradations necessarily must exist between normal and hydatiform villi, thus contradicting Marchand's conclusion.


Menu said that the presence of marked epithelial proliferation was emphasized early by Miiller (1847), Ercolani (1876), Franque (1896), and Owry (1897), and according to Pazzi (1908 b ), Ercolani, and even Polano, denied the existence of connective tissue in the hydatiform mole. The same thing was asserted by Sfameni (1905), who claimed to have found further evidence of the exclusively epithelial nature of the hydatiform mole. According to Sfameni, the hydatiform mole does not result from a modification of existing chorionic villi, but from an entirely new growth which is wholly epithelial in character! But this opinion, which was accepted also by Niosi (1905), seems to exist among Italian writers only. According to Acconci (1914 a ), marked proliferation of the epithelium occurs also in toxemia of pregnancy and in nephritis. A number of investigators have found it common also in long retention.


I am unable to confirm the observation of Nattan-Larrier and Brindeau (1908) that the syncytium of hydatiform villi breaks up into individual portions which do not undergo degenerative changes, but penetrate deeper into the decidua. These investigators thought that in normal villi the plasmodium always keeps its continuity with the proliferating Langhans layer, and that the syncytial masses were more angular in form, of smaller size, and contained retracted nuclei. The syncytial masses on the hydatiform villus, on the other hand, were said to be more rounded in outline, elongated or polygonal in form, and to possess large nuclei very rich in chromatin.


Although Durante (1898) represented extremely long syncytial buds, he nevertheless found (1909) epithelial proliferation present only where certain vascular changes were present. Winter (1907) stated that the condition of the epithelium varies greatly, and Falgowski (1911) emphasized that he could not demonstrate the presence of an increased epithelial proliferation or of vacuolation of the syncytium. Amann (1916) also found that epithelial proliferation may be wholly absent.


That the degree of epithelial proliferation varies greatly, and necessarily so, not only in the villi of the different vesicles but in those of the same vesicle, is splendidly illustrated by the villi of No. 7206, shown in figures 107 and 108. In the former epithelial proliferation has not extended beyond that required by the increase in surface due to the increase in caliber of the villi, while in the latter a very long, branching epithelial framework and smaller processes are present. Likewise, in figure 109 (No. 540), the large villi show little epithelial proliferation, while the small villus to the right shows very marked proliferation. Nor are the variations in the degree of epithelial development limited to different "villi of the same specimen, for they may be present even in the same villus. Ballantyne and Young (1913), on the contrary, found epithelial proliferation "so well developed that it suggested that it is an essential process in the formation of the mole." They further likened hydatiform degeneration to edematous growths and emphasized that both really are epithelial new growths. This opinion is accepted also by de Snoo (1914), who regarded the hydatiform mole as a neoplasm of the trophblast with secondary changes in the stroma.


There is no agreement at present as to whether the epithelial changes are primary or secondary. As is well known, Marchand (1895), and Miiller, Ercolani, and Langhans long before that, regarded the epithelial changes as primary, but most investigators seem to have come to an opposite conclusion. Some share the opinion of Schroeder that hydatiform degeneration points to a stimulus resulting in hyperplasia of the entire chorionic villus. Nor is there agreement as to what the initial changes are. Durante (1909) regarded the presence of vessels with an imperfect endothelial lining and with thick infiltrated walls as the initial lesion in hydatiform degeneration. These changes were noted by him, especially in trunk villi, and epithelial proliferation was most evident where the vascular lesions were most pronounced. Durante further stated that the chain form of the hydatids is due to the fact that the vascular lesions occur at intervals along the villus. Unfortunately, the structure of long hydatiform villi does not confirm such an explanation nor Durante's conclusion that the hydatid cavities within the villi result from dilatation of the capillaries. Many investigators report the early disappearance of the bloodvessels, a phenomenon which some regard as secondary and others as primary to the death of the embryo.


In the course of this investigation a villus with a normal stroma and normal vascularization never was found to have undergone true hydatiform degeneration, but one with a normally active epithelium both Langhans layer and syncytium often was truly hydatiform. That is, it not only was watery in appearance, but also fusiform or globular, even in external form. In fact, Marchand (1895) himself found that "Das Epithel welches die Zotten und ihre Anschwellungen bekleidet zeigt ein sehr verschiedenes Verhalten." Yet even today the feeling on the part of many seems to be that unless a marked hyperplasia of the Langhans layer and of the syncytium is present the condition is not one of hydatiform mole. This position seems to me to be untenable, for, as Marchand himself said, the change in epithelium usually is least in the young villi, and it must be added that it is unrecognizable in the early stages and in young conceptuses. A perusal of the literature descriptive of the actual cases leaves little doubt upon this point, and a careful study of the advent of the earliest recognizable changes in hydatiform mole is absolutely convincing. The earliest recognizable, even if not the incipient, changes occur in the stroma and in the vessels and not in the epithelium. In passing, it may be noted that although Marchand stated that the change in the epithelium is primary, he nevertheless somewhat contradictorily added that the most important fact is the degenerative change in the stroma of the villi.


Although not applicable to what I have come to regard as the incipent changes in hydatiform degeneration, it nevertheless is true that the stroma often, if not always, quite early becomes hydatiform that is, glassy or clear, though not necessarily watery. Moreover, the villous vessels often degenerate or disappear completely at a very early stage.


Various grades of hydatiform degeneration with vitreous stroma and vessels in various stages of disappearance are shown in figures 110 (No. 977), 111 (No. 516), 112 (No. 8746), and 113 (No. 396). The first and last of these specimens are in a splendid state of preservation, so that maceration changes really can be wholly eliminated. It is exceedingly difficult to make any definite statement as to what is typical regarding the epithelium. This has been said by others also. Indeed, this necessarily follows from the fact, agreed to by every one, that histologically there is no true line of demarcation between the ordinary benign hydatiform mole, the so-called destructive benign (?) hydatiform mole, whatever its status may be, and the malignant l^datiform mole or chorio-epithelioma. Such a conclusion alone presupposes the existence of the widest differences in the condition of the epithelium in these cases, and that such differences actually exist is beyond question.


Marchand's revolutionary investigation on chorio-epithelioma notwithstanding, the epithelium is not always two-layered, nor is it always thickened, in hydatiform mole. That the epithelium can not always be active beyond the normal follows also from the fact that the proliferative changes in it are subsequent to, even if not necessarily consequent upon, changes in the stroma. Furthermore, like the latter, they are gradual in their evolution and may stop or be stopped at any stage of their development. Then, too, the condition of the epithelium depends very largely upon the preservation of the abortus, and this, as is well known, varies greatly. Finzi (1908) also found that the epithelium may be perfectly preserved or totally destroyed, and that central degeneration is the most noticeable thing and due to the absence of vessels.

Figure 114
Figure 115

However, the most striking thing about the epithelium usually is not its thickness, the presence of large masses of trophoblast, or of numerous syncytial buds, but its splendid state of preservation, especially as contrasted with that of the stroma. This is true of all except macerated or degenerate specimens, for the life of the epithelium seems assured as long as there are periodic accessions of fresh blood, which, as the clinical histories illustrate, is usually the case. The stroma, on the other hand, probably not being wholly independent of the contained capillaries, is deprived very largely of its sustenance during, even if not in consequence of, their degeneration. According to some, hydatiform degeneration of the stroma is the result of an accumulation of nutritive products in consequence of the absence of the vessels. Degeneration of stroma and vessels, however, may result from malnutrition due to poor implantation. Daels (1908 b ) reported that the stroma was densest where the syncytium was thickest, and most rarefied where it was thinnest, but I could not determine such a relationship.

The epithelium of the villi often was found single-layered without any syncytium whatever, or with at most a few syncytial buds. Nevertheless, both the syncytium and trophoblast very often show evidences of a marked activity not confined to implanted villi or to the epithelium of the villi as a whole, but which extended to that of the chorionic membrane as well. Surprisingly long, complex, syncytial buds, whorls, and festoons, as shown in figures 114 and 115, said to have been observed by Fraenkel, often are present, especially on the villi, although in a few instances fine buds and frameworks of syncytium also were seen arising from the epithelium of the chorionic membrane. This feature (shown in fig. 115) has, I believe, not been specially emphasized heretofore, though observed by Clivio (1908).


Mounds formed by the Langhans layer were common, especially on the tips of the villi where they frequently formed irregular masses of small nodules the "appendici durate" of Crosti (1895). These gave the villous tree the appearance of a leafless orange loaded with fruit, only that the trophoblastic nodules are mainly apical, as shown in figure 116. In several instances syncytial buds were found far out on these trophoblastic masses, a fact which is of special if not of crucial significance in connection with the old question of the origin of the syncytium, for these buds undoubtedly had not been transported there. But, however one may regard these things, such appearances as represented in figure 9 (plate 1, Chapter IV) are unmistakable, for they show thickenings composed of Langhans cells and garlands of considerable length, portions of which are composed of absolutely distinct cells of the Langhans type, as well as other portions composed of syncytium with every gradation between the two. Nor do I believe that the assumption that syncytium can resolve itself into individual cells can be used to deny the implication of these facts.


Although hydatiform villi covered by a single layer of rather small cells of the nature of Langhans cells, sometimes without visible cell boundaries, frequently were seen, villi covered by typical syncytium only were never seen. The single layer present, although syncytial in places, suggested Langhans cells rather than the real syncytium. Moreover, since the cells of the Langhans layer usually were smaller rather than larger than normal, it follows from this alone that their proliferation nevertheless must have been marked, in order to completely cover the enlarged villus, in spite of the fact that the layer remained single-celled. Were this not the case the extraordinary increase in size which accompanies the formation of large hydatid cysts could not possibly occur without rupture of the covering layer.


Not infrequently proliferation of the epithelium without increase in thickness may manifest itself in another way. The caliber of the villi in the earlier stages of hydatiform degeneration sometimes does not increase much and no thickening of the proliferating epithelium is noticeable, yet the latter shows marked proliferation. Under these circumstances, the borders of the villi and of the chorionic epithelium may appear extraordinarily sinuous, as illustrated in figure 117, and epithelial invaginations from opposite sides may in rare instances meet in the center, as indicated in figure 118, and, by fusion, completely isolate a portion of the stroma. It is usually in these cases of very sinuous epithelium that the epithelial invaginations sometimes become constricted, leaving a closed epithelial vesicle or a nodule of epithelium attached to a stalk or wholly isolated within the stroma, as shown in figures 119 and 120. All stages in this process of vesicle formation were found, and rarely also extensions of epithelial sprouts, as described by Neumann (1897) and others, were seen, portions of which had become isolated in the stroma to appear later as typical syncytial giant cells. These facts, too, would seem to throw a sidelight upon the origin of the syncytium for those to whom this question is still an open one.


All these things abundantly testify to the activity on the part of the epithelium in many hydatiform moles, even when thickening of it is absent, but they are of diagnostic value only if present, and I wish to emphasize again that they may be wholly absent and also unrecognizable in the early stages. Moreover, the degree of epithelial proliferation varies greatly, as illustrated in figures 112, 121, and 122.

Until I am able to learn more about the structure of normal villi in various stages of development, I am not willing to commit myself regarding the earliest changes in hydatiform degeneration. These may be unrecognizable with present methods. However, it is possible to say that in young conceptuses the disappearance of the capillaries, which was regarded as a posisible cause for the development of hydatiform mole by Hewitt (1860, 1861), and which was emphasized later by Hahn (1864), Maslowsky (1882), and also by others, undoubtedly is a very early and possibly the very earliest noticeable change in some cases. I do not imply that death of the embryo is the cause of this disappearance, as Hewitt held, and I am not ready to say that the vascular change is the very earliest one in all cases. This would imply that hydatiform degeneration under no circumstances can begin before the capillaries have appeared in the villi. There is some evidence which suggests that it possibly may appear before this time. If so, it would be incorrect to speak of a disappearance of the vessels in such chorionic vesicles, for if the advent of hydatiform degeneration can precede the appearance of the villous capillaries, vascularization of the villi may never occur. In older conceptuses, however, in which vascularization of the villi has supervened, the first recognizable change is the disappearance of these capillaries. Many specimens in which the latter were in various stages of degeneration were examined carefully, and the opinion of Hewitt (1860) that hydatiform degeneration can not arise in villi which have been vascularized can be regarded as of historical interest only. Different stages in the process of vascular degeneration are represented in figures 123, 124, and 126, and in figures 109 to 112, inclusive.

Coincident with the disappearance of the vessels, also noted by Vecchi (1906) in villi with vesicles only as large as a "millet" seed, changes in the stroma also are noticeable. Usually it tends to become glassy, the individual nuclei becoming separated farther. The stroma, though apparently solid, is uniformly slightly bluish and vitreous, with well-defined, rather small, pycnotic, pointed nuclei, but with not a vestige of a vessel, though the epithelium be splendidly preserved. The latter may be one or two layered, and may be accompanied by syncytial buds and trophoblastic masses and nodules. In such specimens the entire picture really is exquisite, and a mere glance through the compound microscope reveals the lack of vessels in the vitreous stroma, its sparseness, and the marked differences in size of the sections of the villi.


After these early changes, liquefaction of the stroma usually follows. As is well known, it generally begins in the interior and first appears in the form of vacuolation; but this vacuolation (which I can not regard merely as an edema) is not intracellular but intercellular, and as it becomes more pronounced it really takes on the nature of fenestration. Sections of the whole cross-section of the villi, even though large, may be composed of a series of fenestrae (see fig. 125) separated by exceedingly fine strands of the remaining stroma which may contain remnants of the nuclei. But finally, even the fine trabeculae separating the fenestrse disappear, and the stage of the watery, old, hydatid condition has been reached. More generally, however, the vacuoles or small fenestrs lying in the middle become confluent at the center of the cross-section of the villus, which then is liquefied completely. As is well known, this liquefaction gradually extends to the periphery as the zone of the surrounding stroma is narrowed in the process. Not infrequently, however, liquefaction of the stroma occurs quite generally throughout the crosssection of the villus and is accompanied by the formation of numerous large cells the wandering or migrating cells of earlier writers. A few of these cells almost always can be found, and rarely the whole section of the villus is studded with or even formed by these large, erratic cells, which usually lie in fenestra? in the stroma (fig. 127.) In other instances a large portion of the sections of the villi may be occupied by them, as shown in figure 128. The presence of these cells in villi regarded as normal has long been known. Their presence in hydatiform moles was noted by Otto, Marchand (1898), Essen-Moller, and many others. Their occurrence in normal and pathological chorionic vesicles and their significance are considered in Chapter XV. No matter what the condition of the epithelium (or, more specifically, the Langhans layer), the syncytium, and trophoblast may be, the above-noted changes in the stroma always are quite typical. They are not the only changes noted, however, and may differ somewhat as to the time of their advent.


Not infrequently, changes quite comparable to those in the villi occur also in the stroma of the chorionic membrane itself, a fact which has not heretofore been emphasized. It, too, frequently is decidedly glassy; liquefaction may occur here and there and may become complete in the course of time. Hofbauer cells also are not uncommonly present. Among the changes noted in this membrane, the disappearance of the vessels is the most common and constant, although epithelial proliferation is not rare, as already stated. Moreover, when (as in one of Storch's cases) a hydatiform villus is 15 cm. long, one scarcely can doubt that the stroma also must have proliferated not merely degenerated. Some of the strings of hydatid cysts in a specimen in the Carnegie Collection have a length of 10 to 12 cm., and in these cases also one can hardly assume that this increased length of the villi was unaccompanied by proliferation of the stroma. From these things alone it follows that the stroma can not remain passive always, although Gromadzki (1913) concluded that it never proliferates. Vecchi (1906), however, reported an increase in the stroma of the villi, and it will be recalled that Marchand also implied the presence of proliferative changes in the .connective tissue when he wrote that they depend upon those in the epithelium.


I have never been able to find mitotic figures, a fact which may be accounted for, however, by the presence of degenerative changes due to intrauterine separation and retention of most specimens. Indeed, the failure to find mitoses speaks against proliferation in the stroma no more than in case of the epithelium, in which the presence of karyokinetic figures has been reported by a few investigators only. Yet pronounced proliferation of the epithelium is often present. The failure to find mitotic figures very likely is due to the condition of the material.


Careful scrutiny of a large series of specimens has revealed the fact that the disappearance of the vessels in the villi, in the chorionic membrane, and also in the umbilical cord is centripetal as a rule. However, in many specimens the vessels not only may be present in the chorionic membrane, although absent in the villi, but may be very numerous and even engorged with blood. It is difficult to say to what extent the engorged condition of these vessels and of those in the body of the abnormal embryos sometimes contained in these hydatiform moles is due to the propulsion of the contained embryonic blood before the advancing vascular constriction and degeneration, but I am inclined to believe that the centripetal movement of the process is not a negligible factor in this matter.


Although only a few instances of the birth of a living fetus or of a fetus which has reached the later months of pregnancy are recorded in the literature, it is now quite generally recognized that the fetus, though dead and too small for its menstrual age, usually is present. This stands in contradiction to the earlier belief illustrated by the statement of Gierse (1847), that the fetus usually was reported as absent, and that when present (as in the cases of Meckel, Gregorini, Otto, Cruveilhier, and his own) it was usually less than an inch long, even when retained for a period of from 3 to 10 months.


This apparent contradiction regarding the presence of the fetus in hydatiform moles is explained easily by the fact that the cases in the earlier literature are old, far advanced in degeneration, while the more recent literature contains many more in the earlier stages of degeneration. Yet in spite of this fact the earlier opinion survives to the present day, for Graves (1909-10) spoke of "the very unusual presence of a normal fetus inside a mole," and Vineberg (1911) still more strangely held that the presence of a fetus excludes the specimen from the class of true hydatiform moles!


Among the specimens concerned in this report many contained a fetus. This was true of 24.5 per cent of 49 tubal and 64.4 per cent of 121 uterine specimens, including some (9) doubtful cases. The fetal length ranges from 1 to 90 mm. in the uterine and from 1 to 80 mm. in the tubal series. Although the average length of the embryo in the tubal series is 12.3 mm., and that of the uterine only 10.1 mm., 58 per cent of the tubal specimens nevertheless were below 7 mm. in length as contrasted with 52.5 per cent of the uterine. The presence of a fetus with a frequency almost three times as great in the uterine series again indicates that the abnormal conditions within the tubes lead to early death, digestion, and absorption, or at least to dissolution, of the embryo. This fact again points directly to a faulty nidus as causative agent, for if the absence of a fetus is to be laid to primary ovular defects, then one must admit that relatively far more of such diseased ova become implanted within the tube than within the uterus.


In some early specimens the fetus is in a state of excellent preservation. This is what one might expect, for the onset of hydatiform degeneration is gradual and may remain partial. The condition of the fetus alone in many of them also suggests that its death was secondary to the degeneration.

Figure 129
Figure 130

This is exemplified splendidly by No. 2099, shown in figures 129 and 130. This cyema shows the presence of undoubted maceration changes throughout, but especially in the branchial region and on the umbilical cord, and the shaggy chorionic vesicle shows the presence of a moderate degree of hydatiform degeneration. A more advanced instance is that of No. 1260, shown in figures 131, 132, and 133, in which the form of the cyema is greatly modified and the hydatiform nature of the villi much more obvious. That the death of the embryo is not the cause of hydatiform degeneration would seem to be evident also from such instances as No. 2250. In this case of twin pregnancy both fetuses are well preserved, as figure 134 shows, and yet the respective chorionic vesicles show the presence of welldeveloped and quite general hydatiform degeneration as can be seen in figures 141 and 142 (plate 14, Chap. X). Indeed, in this instance the fetuses show less maceration change than the vesicles, and it is particularly interesting that this abortion was attributed to influenza, although, in view of the presence of hydatiform degeneration, it was inevitable. Influenza, to be sure, may have been the immediate cause and may have precipitated the abortion, but uninterrupted development of the hydatiform degeneration also would have done so. Moreover, the appearance of both fetuses suggests that they died shortly before the abortion, and this is confirmed by a comparison of the menstrual and anatomical ages, which differ by 6 and 10 days, if the larger or smaller fetus is used for the determination of the latter age. However, if we can finally assume that the menstrual age of all conceptuses exceeds their true age by about 10 days, then they must have lived up to the tune of abortion, or, strictly speaking, a few days beyond it. That death of the cyema is not the cause of hydatiform degeneration is indicated also by such specimens as No. 2411, represented in figures 143 and 144 (plate 14, Chap. X), a twin, double-ovum pregnancy, in which both the cyemata and the chorionic vesicles show considerable and apparently the same degree of maceration (figs. 145 to 147). Since both chorionic vesicles also show the presence of quite general hydatiform degeneration, it is evident that if the latter had arisen only after the death of the cyemata the vesicles should not show anything like a corresponding degree of maceration, unless perhaps the time of retention had been considerable.

Many other specimens of single pregnancy could be used to illustrate the same thing, and since the development of hydatiform degeneration undoubtedly is not a fulminating one, it might be expected that considerable development of it might occur before the death of the cyema, which is due apparently to the obliteration of the villous circulation. Since blood-vessels can and do arise in the chorionic villi quite independently of those in the cyema, it also seems possible that young chorionic vesicles showing hydatiform villi with disappearing bloodvessels may be found, even if the intra-cyemic circulation never developed or never united with the extra-cyernic or chorionic circulation. Such a surmise does not imply, however, that hydatiform degeneration never begins before the bloodvessels appear.


Of the many explanations which have been offered for the advent of hydatiform degeneration, none seems to be better established than that of endometritis. This was first emphasized by Virchow (1863). Lwow (1892) also reported 4 cases in patients under his care in whom lues could be excluded, and in whom he held endometritis responsible. Emanuel (1895) was the first, it seems, to demonstrate the presence of cocci in inflammatory foci of round cells in the decidua accompanying a case of hydatiform mole. Veit (1899) also believed that disease of the decidua is the cause of hydatiform degeneration. Veit further stated that Waldeyer, Jarotzky, and Storch also believed that an irritative condition of the decidua is responsible. Stoffel (1905) also found cocci other than gonococci present, and says he can not avoid holding endometritis responsible in his case. The association of hydatiform degeneration and endometritis was noted also by Marchand (1895), Oster (1904), and Sternberg; also by Essen-Moller, who reported the phenomenal case of a woman with endometritis, who had aborted a hydatiform mole 18 times in 9 years. Falgowski, on the contrary, concluded that the ova themselves were diseased and argued that hydatiform degeneration should be much more common if it were due to endometritis. Taussig (1911) also stated that leucocytic infiltration of the decidua is frequently present in hydatiform moles, but insisted that "leucocytic infiltration in the placenta then should not be interpreted as infection. . . . Inflammation and infection should be kept apart." I presume Taussig really meant infiltration and infection should be kept apart, and the question then turns upon the structure of the normal decidua and the significance of infiltration for the development of the ovum.


It may be recalled that Marchand (1904) reported the presence of isolated groups of small cells in the normal decidua which looked like mononuclears under low magnification, and which he believed have often been confused with them. But even granting this, and the further facts that the exact histologic changes in the decidua are not fully known, and that it is rather difficult to ascertain just what decidual changes are regarded as evidence of the existence of an endometritis, any one examining a large series of cases of hydatiform degeneration aborted with the decidua can not doubt the presence of marked decidual changes in a very large percentage of them. These changes are not limited to infiltration with scattered round cells or erythrocytes, or to focal accumulation of the same, but often extend to almost complete fibrosis, as shown in figure 135, plate 13 (see Chap. IX), so that experienced investigators have mistaken the thin, fibrous decidua for a part of the chorionic vesicle.


It is true that the existence of these changes in the decidua does not necessarily imply that they were antecedent to the implanation of the ovum, but fortunately the clinical histories and material from curettage often supply crucial evidence. From such cases and from the cumulative weight of evidence from the large series of cases here reported, the great majority of which showed decidual infiltration or other changes suggestive of endometritis, the frequent association of abnormal deciduse with hydatiform degeneration is evident. The fact that the incidence of hydatiform degeneration in the tubal was considerably higher than that in the uterine series might be regarded as contradicting this relationship, but such is not the case. The mucosa of the tubes at best is an unfavorable nidus for implantation because of the absence of decidual formation alone. Hence, even if salpingitis were somewhat less frequent than endometritis, difficult nidification in the tube could easily more than account for the existing differences. Hence the higher incidence of hydatiform degeneration in the tubal series in fact becomes confirmatory of the conclusion that abnormal nidification really may be responsible for the advent of hydatiform degeneration.


The only fact which might be interpreted as indicating that germinal defects primarily are responsible for the development of hydatiform degeneration is the relatively higher incidence of the condition in older women. Against this, however, stands the other fact that such women also show the cumulative effects of endometritis and pregnancy upon the endometrium. Furthermore, since hydatiform degeneration so often follows one or two normal births or abortions, it would be impossible to find an adequate explanation for the release of the defective ova so often after and not before these events.


I am reminded also in this connection of a case the detailed history of which is fully known. It is that of a robust young woman who successively gave birth to two moles and then to a normal full-term child and secundines. In this case curettage was done in connection with each mole. Apparently the new endometrium, which had formed after the second abortion and curettage, permitted normal implantation and normal development to progress to term. To ignore the condition of the endometrium in this case and attribute the development of hydatiform degeneration to the successive releases of abnormal ova would seem to disregard important facts especially so since no one has established the occurrence of abnormal ova within the Graafian follicle, a possibility which I do not wish to deny, although Donskoj's report of a case of hereditary mole must surely be taken cum grano salis.


That an abnormal nidus may be responsible for the advent of hydatiform degeneration would seem to be indicated also by the fact that the process usually was better developed and more general in the tubal than in the uterine cases. That both endometrium and decidua show astonishing differences in structure under pathologic conditions is well known. The entire tubal mucosa, on the other hand, even when normal, forms an abnormal nidus which would affect all portions of early chorionic vesicles somewhat alike, and since, as found by Mall, inflammatory conditions in the tubes predispose to tubal implantation, the higher incidence of hydatiform degeneration in the tubes is easily explained. Nor does the existence of partial hydatiform degeneration argue against such an explanation.


Although Kehrer reported not a single fatality in 50 cases of hydatiform mole, Hirtzman (according to von Winckel) gave the fatality as 13 per cent, Borland and Gerson as 18, and Williamson as 20 to 30 per cent. Von Winckel (1904) regarded these percentages as entirely too high, however, although Oster (1904) reported 2 cases of malignancy out of 15 cases in which the late results were ascertainable. This is an incidence of 13.3 per cent. Since none of 15 patients who had suffered from hydatiform mole had a recurrence, Oster concluded that the cause of hydatiform degeneration is a fortuitous one. Schickele (1906) stated that only 3 cases of hydatiform mole with coincident chorio-epithelioma were found in the literature; and Nattan-Larrier and Brindeau (1908) seem to avoid the implied difficulty by premising the existence of a histological as contrasted with a clinical malignancy, a distinction which recalls the conclusion of Schickele that it does not follow that not all hydatiform moles are malignant merely because not all of them cause the death of the patient. Kroemer (1907) found that chorio-epithelioma developed in 5 out of 15 cases of hydatiform moles, or in 33.3 per cent, but only twice in 3,841 "normal implantations." Daels (1908) says La Torre claimed a malignancy of 64 per cent, de Senarcleus one of 28.7 per cent, or 14 out of 49 cases. L. Fraenkel (1910 a ) emphasized that the estimates of the number of cases in which hydatiform degeneration is followed by malignant disease vary greatly, while Robertson (1915) quoted Findlay as finding that 16 per cent of 250 hydatiform moles collected from the literature were followed by malignant disease. Briggs, who reported 21 cases of hydatiform degeneration with 2 of chorio-epithelioma, or an incidence of malignancy of 9.5 per cent, called attention to the "diminishing ratio in the tendency to malignancy" shown by his series.


Findlay (1917) stated that chorio-epithelioma developed in 131 out of 500 cases gathered by him from the literature, which is an incidence of 26.2 per cent, but, as already stated, most of these cases from the literature are old, advanced degenerations, many of which have been retained for a long time. The tendency to malignancy in these probably can in no way be compared to that in smaller and younger specimens, many of which are aborted entire with the surrounding decidua. Consequently it need not surprise us that out of 19 cases of this series, in which later reports were obtainable, none was reported as having developed chorioepithelioma.

Perhaps I may add a word of caution in regard to a possible change in attitude toward the question of malignancy with a consequent relaxation of vigilance. It is true that out of the 21 cases of Briggs only 2 developed chorio-epithelioma; but it must not be forgotten that Briggs in part was, and I to a far larger extent am, dealing with a different class of hydatiform moles than those upon a study of which the prevailing conception of malignancy is based. Hydatiform moles which continue to grow and which survive for months after the death of the embryo evidently are more vigorous, and hence no doubt also more dangerous than those which are aborted early and spontaneously. Since the latter formed the great majority of all moles here considered, opinions regarding malignancy formed on this basis probably would lead to disaster if applied in practice. Such conceptions would be based upon a totally different incidence than the current one of 1 hydatiform mole in every 2,000 cases. Instead of relaxing our vigilance, it would seem wise to increase it, particularly in the cases of so-called spontaneous abortions the cases in which no ascertainable cause for the termination of pregnancy can be found, especially if the chorionic vesicle is empty or if the embryo belongs in one of the early groups of Mall's classification.


The average age of 36 women aborting hydatiform moles was 31 years. Although I do not regard the alleged ages as necessarily the actual ones, this average age agrees very well with that of 6 cases reported by Poten, 10 by Donskoj, 23 by Briggs, 6 by Gromadski, and 8 by Robertson. The average age of Poten's cases was 32 years, of Donskoj 's 25 years, of Brigg's 28 years, of Gromadski's 29.6 years, and of Robertson's 28.4 years. Pazzi (1908 b ), on the other hand, stated that Briquel placed the greatest frequency of hydatiform degeneration between 20 and 30 years. These averages are so far on the near side of the menopause that one can make liberal allowances for the proverbial disinclination of women to state their exact age, even to physicians, and nevertheless regard the prevailing opinion regarding the greater frequency of hydatiform mole near the menopause undoubtedly as ill-founded. If, as Lewis and Lewis (1906) stated, it is necessary to add only half a year to the average age of a large group of women in order to ascertain the actual average age when considering general social statistics, then everyone will admit that still less allowance than this need be made in the case of women who are speaking to their physicians, knowing that whatever they may say will be regarded as strictly confidential. That it is unnecessary to make large allowances for understatement of their age on the part of these women is indicated also by the average duration of their married life before aborting moles. This in the case of 29 women was 7.1 years. Hence, if one bears in mind that the average age of first marriages, according to Webb (1911), is 25.1 years, one easily can see that the average age of the women aborting hydatiform moles, which was given as 29.6 years, probably is not too low at all, thus confirming the findings of Williamson, who denied that hydatiform mole was especially common near the menopause.


The conclusion that the average age of 29.6 years undoubtedly is near the actual is confirmed also by the fact that a hydatiform mole was the first abortion in 19 out of 41 women, or almost half the number; 12, or almost one-third, had aborted twice; and only 10 had aborted more than twice. But what is still more confirmatory is the existence of a surprising parallelism between the data on abortion and those on births; 9 of 33 women had given birth to but 1 child, and an equal number had given birth to but 2. Hence over 50 per cent of the 33 women had borne children twice, or less than twice, and only 15, or less than half, had borne oftener than this.


This undoubted evidence of the youth of these women is confirmed still further by the statement of Lewis and Lewis, who, from an analysis of 16,325 first births, found that nearly one-half of them occur between the ages of 20 and 24, and almost three-fourths between 20 and 29 years, although first births are more frequent between 30 and 40 than between 15 and 19 years. I realize that social statistics can not be translated from one country to another without modification, but in such a mixed population as ours this modification probably need be less, rather than greater, than in case of some countries.


The conclusion that the occurrence of but a single birth before the advent of hydatiform degeneration probably implies that such women are relatively young is emphasized still further by the statement of Lewis and Lewis that in one-third of the marriages in Scotland "the bride had a child when unmarried or was pregnant at the time of marriage," and that 50 per cent of the first births in Scotland occur within 9 to 24 months after marriage. Lewis and Lewis also give the average interval between marriage and the first birth in 16,176 cases as 13.54 months, but little more than one year. Since Lewis and Lewis stated that the interval between the birth of the first and that of the second child is but little longer than that between marriage and the birth of the first child, being only 3.07 years, it is evident that not even those women who had borne two children before the advent of hydatiform degeneration could have been near the menopause. This conclusion is emphasized still further by the fact that in 96.12 per cent of 16,176 fruitful marriages fertility was demonstrated within three years after marriage.


Nevertheless, in spite of the clear implication of all these facts, I wish to emphasize again that since what have been heretofore regarded as hydatiform degenerations were large specimens mainly, it well may be, and according to certain authors it is true, that such cases occur later in the reproductive life of women. Yet it certainly is significant that Findlay, in tabulating 500 such cases from the literature, found that 275, or 55 per cent, occurred before the thirty-fifth year, and of 36 specimens from the Carnegie Collection 23, or 63.6 per cent, came from women below this age. It may also be recalled that 78 per cent of Kehrer's 50 cases and 90 per cent of Bloch's occurred before the fourth decade. That hydatiform degeneration may occur very early in life is suggested by the remarkable case of Strieker (1879) in a precocious child of 9 years.


Fourteen out of 23 cases, or 61.3 per cent of the uterine series, in which the age was given, occurred at or before the thirtieth year, and 18, or approximately 80 per cent, at or before the thirty-fifth year. These things abundantly emphasize the conclusion, reached by some investigators, that hydatiform mole is not absolutely more common at or near the menopause; but it nevertheless may be relatively more common. That is, the number of hydatiform moles aborted after 40, compared with the total number of pregnancies or births after 40, actually may be greater than this ratio before 40 years.


From calculations based on data given by Lewis and Lewis, the average number of births occurring after 40 years in Sweden, Norway, Denmark, Brunswick, Berlin, Budapesth, France, and Scotland is 9.9 per cent. This agrees remarkably well with Bloch's estimate of 10 per cent. But if 77.2 per cent of the cases of hydatiform mole occur below 40, and 22.8 per cent after that year, then it is evident that hydatiform mole nevertheless is relatively more common after than before 40 years, for approximately one-fourth of the cases of hydatiform degeneration would be associated with one-tenth of the births. This would be an increased frequency of 300 per cent after 40 years. A similar result would be obtained by comparing Findlay's or Williamson's series. Hence, hydatiform degeneration, though absolutely less, is relatively more frequent in later life. This fact, however, does not necessarily imply that age in itself is responsible for the increased incidence after 40. A comparison of the incidences of hydatiform degeneration in young and old primipara3, of good health, might elucidate this question.


These statistics are not in agreement with the prevailing opinion that hydatiform moles are more common in multipart than in primiparse. Indeed, they suggest rather that after the first conception, which was normal in a large percentage of these young women, something happened which interfered with the normal development of succeeding conceptions. That, it seems to me, is extremely significant and very suggestive. Here is a group of relatively young women, over 50 per cent of whom had borne but twice and some only once, and then gave birth to a hydatiform mole. While I realize the necessity for circumspection, especially in these matters, these facts seem to me to suggest that something happened to a normal endometrium. Other facts also point in the same direction.

Description of Plates

Plate 8

Mall Meyer1921 plate08.jpg

Fig. 96. Young chorionic vesicle in situ, showing hydatiform degeneration of the stroma and syncytial budding. No. 836. X2.6.

Fig. 97. Same, under greater enlargement, with the cyema in view. X3.75.

Fig. 98. Hydatiform degeneration. A small portion of No. 1189 still in loco. X4.5.

Fig. 99. Another portion of the same specimen. X22.5.

Fig. 100. External appearance of an apparently normal conceptus, partly surrounded by decidua. No. 2077. X0.75.

Fig. 101. A portion of the same specimen, clearly showing hydatiform degeneration where in focus. The decidua is reflected. X3.

Fig. 102. Portion of villi from No. 690, showing numerous syncytial buds, absence of vessels, and glassy stroma. X37.5.

Fig. 103. A chorionic vesicle showing shaggy appearance. No. 2258. XI. 5.

Fig. 104. Hydatiform villi from same specimen. X4.5.

Fig. 105. Macerated normal villi in blood clot. No. 640. X71.

Fig. 106. Well preserved, early hydatiform villi from the same specimen. X33.25.

Fig. 107. Hydatiform villi without marked epithelial proliferation. No. 7206.


Plate 8: Fig. 96 | Fig. 97 | Fig. 98 | Fig. 99 | Fig. 100 | Fig. 101 | Fig. 102 | Fig. 103 | Fig. 104 | Fig. 105 | Fig. 106 | Fig. 107 | Chapter 8. Hydatiform Degeneration in Uterine Pregnancy

Plate 9

Mall Meyer1921 plate09.jpg

Fig. 108. Hydatiform villi with pronounced epithelial proliferation. No. 7206, also shown in figure 107. X50.

Fig. 109. Hydatiform villi without (at left) and with marked epithelial proliferation (at right). No. 540.

Fig. 110. Fairly late stage in the degeneration of the villous vessels in hydatiform villi. No. 977. X50.

Fig. 111. A slightly later stage in the degeneration of the villous vessels. No. 516.

Fig. 112. Almost complete disappearance of the vessels. No. 8746. X50.

Fig. 113. Non-vascular, early hydatiform implanted villi. No. 396. X50.

Plate 9: Fig. 108 | Fig. 109 | Fig. 110 | Fig. 111 | Fig. 112 | Fig. 113 | Chapter 8. Hydatiform Degeneration in Uterine Pregnancy

Plate 10

Mall Meyer1921 plate10.jpg

Fig. 114. Portion of No. 435a, showing a small villus in the center with an extremely long syncytial bud. X50.

Fig. 115. Framework of syncytium arising from the chorionic membrane and from the villi. No. 962<z. X50.

Fig. 116. Apical trophoblastic nodule. No. 516. X6.

Fig. 117. A portion of the ehorionic membrane from No. 714, showing decidedly sinuous epithelium. X50.

Fig. 118. An early, non- vascular, hydatid villus from the same specimen, showing constrictions. X50.

Fig. 119. Epithelial vesicles within the stroma. No. 872. X180.

Fig. 120. Extremity of an epithelial vesicle within the stroma, same specimen. X300.


Plate 10: Fig. 114 | Fig. 115 | Fig. 116 | Fig. 117 | Fig. 118 | Fig. 119 | Fig. 120 | Chapter 8. Hydatiform Degeneration in Uterine Pregnancy

Plate 11

Mall Meyer1921 plate11.jpg

Fig. 121. Slight epithelial proliferation on markedly hydatiform villi. No. 134. X50.

Fig. 122. Extremely marked epithelial proliferation on a small hydatiform villus with glassy stroma. No. 415. X95.

Fig. 123. The disintegrating capillaries are represented merely by large, incomplete curved outlines. No. 749. X50.

Fig. 124. Collapsed capillaries in process of obliteration. No. 712. X95.

Fig. 125. Fenestrated, macerated, hydatiform villi among others which are quite fibrous. No. 651d. X50.

Fig. 126. Villi showing only a trace of the capillaries. No. 651<7- X50.

Plate 11: Fig. 120 | Fig. 121 | Fig. 122 | Fig. 123 | Fig. 124 | Fig. 125 | Fig. 126 | Chapter 8. Hydatiform Degeneration in Uterine Pregnancy

Plate 12

Mall Meyer1921 plate12.jpg


Fig. 127. Section of a villus from No. 510, showing scattered Hofbauer cells. X300.

Fig. 128. Section of a villus from No. 749, showing accumulation of Hofbauer cells, especially at the right. X50.

Fig. 129. A somewhat macerated normal embryo from a case of early hydatiform degeneration. No. 2099. X6.

Fig. 130. The chorionic vesicle of the same specimen, showing early hydatiform villi to left where in focus. No. 2099. X4.

Fig. 131. Basal area of No. 1260, showing pronounced hydatiform degeneration of the villi. XI.

Fig. 132. The same specimen, opened in the capsular region to show the presence of the cyema. No. 1260 Xl.l.

Fig. 133. Marked macerated and deformed fetus from the same case. No. 1260 X4.

Fig. 134. Twin fetuses, No. 2250a and 6, a case of hydatiform degeneration. XI. (See also figs. 141 and 142, plate 14.)


Plate 12: Fig. 127 | Fig. 128 | Fig. 129 | Fig. 130 | Fig. 131 | Fig. 132 | Fig. 133 | Fig. 134 | Chapter 8. Hydatiform Degeneration in Uterine Pregnancy


Embryology - 13 Dec 2018    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)

Mall FP. and Meyer AW. Studies on abortuses: a survey of pathologic ova in the Carnegie Embryological Collection. (1921) Contrib. Embryol., Carnegie Inst. Wash. Publ. 275, 12: 1-364.

In this historic 1921 pathology paper, figures and plates of abnormal embryos are not suitable for young students.

1921 Carnegie Collection - Abnormal: Preface | 1 Collection origin | 2 Care and utilization | 3 Classification | 4 Pathologic analysis | 5 Size | 6 Sex incidence | 7 Localized anomalies | 8 Hydatiform uterine | 9 Hydatiform tubal | Chapter 10 Alleged superfetation | 11 Ovarian Pregnancy | 12 Lysis and resorption | 13 Postmortem intrauterine | 14 Hofbauer cells | 15 Villi | 16 Villous nodules | 17 Syphilitic changes | 18 Aspects | Bibliography | Figures | Contribution No.56 | Contributions Series | Embryology History

Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic Textbook" and "Historic Embryology" 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 and interpretations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)