Book - Contributions to Embryology Carnegie Institution No.56-3
Chapter 3. Classification
- Uncredited quotations are from notes or circulars written by Mall.
In the now historically very interesting "Prolegomena of the development and metamorphosis of the human ovum" contained in the excellent "Graphic Illustrations of Abortion" published by Granville in 1834, some attempt was made to group the abortuses there reported. The first 6 specimens are spoken of as "lanuginose" ova of several types; the next 3 are designated as denuded or diaphanous; a few are spoken of as opaque; one as "uviform," that is, hydatiform; and several as moles of various kinds. Most of the rest of the 38 specimens were given long individual descriptive names, although several were spoken of as "coriaceous" and others as "avellanated." A case of tubal and the well-known case of ovarian pregnancy are among these specimens.
In speaking of the uviform ovum, Granville emphasized that hydatiform degeneration is a very rare occurrence and gave a very good short characterization of it, saying that although, in this instance, three months had elapsed since pregnancy supervened, nature evidently had spent most of this time "playfully modelling, forming, and cutting out what would appear an artificial plaything, so fantastical it looks."
Panum (1860), from a study of collections of monsters found in various European laboratories, and also from extended experimental work with eggs, formulated a more elaborate classification of deformities, but did not concern himself particularly with the routine material from abortions. It may be recalled, however, that he also spoke of amorphous and cylindrical chick embryos, and that these groups of Panum's practically were equivalent to the nodular and cylindrical embryos of His, Giacomini, and Mall. Panum's group of monstruositates Males cylindrical among abnormal chick embryos were described as young specimens with formless heads and somewhat deformed, rather rod-like bodies, which he ascribed to lowering of the incubator temperature.
His (1882), while considering only younger abortuses, did not recognize the occurrence of empty chorionic and amnionic vesicles, saying that, aside from the two or three specimens in which these cavities were filled with solid blood-clots which made the finding of an embryo impossible, he never saw vesicles devoid of an embryo. This observation is particularly interesting, because empty chorionic vesicles, after all, are not very uncommon and one can hardly escape the conclusion that his failure to see them was due to the fact that physicians sent only the specimens containing embryos. This is not at all unlikely, for they knew, of course, that His's interest centered so very largely upon the form of early human embryos.
All younger embryos abnormal in form, below 15 mm. in length, were classified by His as nodular, atrophic, and cylindrical. Sometimes he also spoke of some of the atrophic forms as flexed or kinked because of the marked drooping of the head upon the chest noticed in some of them. Since these three types of early embryos seem to recur, His emphasized that it was very probable that normal development can be disarranged easily at certain definite stages in prenatal life, a surmise confirmed by contemporary experimental embryology. In the opinion of His, this faulty development almost invariably led to abortion at the end of the second or the beginning of the third month, which period had been noted by A. Hegar (1863) as the time of most frequent abortion of early specimens. It is striking, indeed, that not a single abortus from the later months of pregnancy, which His regarded as capable of continued growth, ever came to his attention.
Giacomini (1888, 1893) also described several nodular atrophic embryos and very significantly stated that he never met with nodular or atrophic forms in rabbits. In these animals he found only advanced deformities, and these he regarded as often merely being the result of death of the fetus with subsequent retention. Hence he did not consider them at all comparable to those found in man. Giacomini also stated that he could experimentally produce all forms of nodular and cylindrical embryos in rabbits, and believed that the atrophic forms were due to changes after death, although he preferred not to emphasize this opinion.
Later, when Giacomini studied these anomalous forms microscopically, he concluded that a classification based on gross appearance only really tells nothing about the nature of the process which has inhibited development, of the time when the changes began, or of their localization and distribution. Hence he emphasized the necessity of a microscopic examination for purposes of scientific classification. He further stated that both the flexed (or kinked) and the cylindrical forms of His could be classed under the atrophic and declared that abortuses can be placed into two main groups, those with and those without an embryo. In 1895 Giacomini further suggested two subdivisions under each of these groups. He divided those without an embryo into (1) specimens with only the embryo missing, and (2) those with little else than the chorion present. Since Giacomini believed that the embryo and even the amnion can migrate from the chorionic vesicle and the hole through which they pass close again by healing, he further differentiated specimens in which the embryo had migrated out entirely, those in which it had reached the coelom externum, those in which the embryo and amnion had migrated from the chorion, and finally, those in which the active migration of the embryo and amnion had everted the chorion. Since the idea of an active migration of the embryo, not to mention that of some of the adnexa, from out the chorion, does not hold, these four types of empty chorionic vesicles require no further consideration. Specimens in which embryos were present Giacomini classed as nodular and atrophic, after His, the latter class including both the cylindrical and the flexed forms of His. Giacomini's great merit lay in recognizing a group of specimens without an embryo, in making a careful microscopic study, and especially in undertaking experimental work in order to elucidate some of the problems which presented themselves to him in a study of abortuses. These aspects of his work will be referred to again in later chapters.
Although other anatomists, and also some clinicians like Giacomini, took up the problem suggested particularly by the work of His, no one followed it with more devotion that Mall. In saying this I am not ignoring the able and extensive treatises on teratology, such as those of Ahlfeld, Ballantyne, Dareste, Saint Hilaire, Schwalbe, and Taruffi. These treatises are more specifically concerned with the problems presented by malformations, while His, Giacomini, and Mall gave especial attention to questions presented by all abortuses, and Giacomini and Mall especially to such as Mall later grouped as pathologic. I realize that I thus raise the question of the relation of uterine pathology to teratology, but that is unavoidable. Moreover, as everyone knows, it is still decidedly unsettled. This attitude was taken also by Mall (1908) when he wrote:
" Whether the early pathological embryos are young monsters of so extreme a degree that they will not continue to grow, is now the most important question of the capital problem in teratology."
In this monograph Mall did not attempt to classify the 163 pathologic embryos collected during a period of 15 years. Nevertheless, as his collection grew, some classification became highly advisable. In referring to this matter later, he wrote:
" In considering a large number of specimens from a scientific standpoint, their classification 1 is one of the most important things. All of our specimens are placed in one of two great divisions normal or pathological. As stated before, the normal ones usually are easily segregated because, for the sake of expediency, we have decided to regard as normal all embryos and fetuses which are normal in form, although they may be inclosed in diseased chorions. These normal specimens are first arranged according to size and subsequently also according to age."
All normal embryos and fetuses further are placed in one of three grades, according to the preservation of their body form. The basis of this gradation is wholly gross morphologic. Moreover, it would perhaps be more correct to say that the deciding factor is not the degree of normality of form which may once have existed, but the extent to which it is preserved. If the form is preserved perfectly, the embryo or fetus is placed in the first of these grades unless it has a localized anomaly. However, abnormal specimens also are found among the group of normals, and, as already indicated, some of the normal specimens no doubt have developed upon a somewhat diseased endometrium, but so far we have been unable to recognize these in the early stages of development.
All specimens the form of which is excellently preserved are put into grade 1 of the normal, but if the external form is changed slightly, either by external forces, by maceration, or by developmental abnormalities, they are put into grade 2. All other specimens regarded as normal in form, with or without anomalies, fall into grade 3. Abnormalities are noted especially by cross-references, but no separate category of abnormal embryos is maintained. Hence, under this plan, a double monster, such as No. 2107, with six extremities, some of which also are monstrous, with one set of deformed genitalia, and with totally fused trunks, becomes classified
1 For the earlier classification of the specimens in the Mall Collection see Mall (1900, 1903, and 1904).
44 STUDIES ON PATHOLOGIC OVA.
as a normal fetus with localized anomalies. Nor is provision made for such specimens in the pathologic group. Since an embryo or fetus may be entirely, or at least approximately, normal in form, even if somewhat macerated, it follows that macerated specimens also are found in these grades of the normal. But, as will appear later, maceration soon changes the form of the embryo. Hence it follows that none except perhaps extremely small specimens can be markedly macerated and still apparently retain their normal form. In these the maceration which is present may become evident upon microscopic examination only.
None of the specimens in grades 2 and 3 is perfect in form, and these grades contain relatively more macerated specimens than grade 1. A splendid example of a normal, slightly macerated fetus is No. 782; although obtained at hysterectomy, it undoubtedly is macerated, yet in spite of this fact it is found in grade 1. This fetus probably died some time before operation, perhaps in cohsequen.ce of endometritis. The embryonic tissues are markedly macerated in spite of the fact that the entire uterus was placed in formalin immediately. Since the chorionic cavity was not infected, this maceration must have taken place under absolutely aseptic conditions, yet it nevertheless is quite evident.
A series of embryos that really does not belong within these two grades of normal specimens are those with localized anomalies. These, of course, really are abnormal, though not necessarily pathologic in the sense of being themselves diseased. They may, however, be the products of disease elsewhere or have arisen from defective ova or spermatozoa. The division of normals also contains specimens not perfect in form merely because they have sustained fractures or have been mutilated, and since no separate group is maintained for those possessing hereditary abnormalities, some of these too are included among the normal.
Specimens which are not included in any of the above three grades of normal embryos or fetuses are classed as pathologic. This does not imply, however, that they themselves necessarily are diseased. In fact many (most?) of these specimens probably were the victims of accident, interference, or untoward conditions, and hence may have been entirely normal at the time of death. In this division also are found empty normal chorionic vesicles from which the embryo was expelled after the membranes had ruptured, or in which the embryo had become disintegrated in consequence of maceration changes due to interference or to other causes. It also contains specimens of fetuses which, for one reason or another, died after the escape of the amniotic fluid, and which then were retained for some time in utero after being expelled from the chorion. The same thing is true of some normal specimens which died in consequence of degenerative changes in the chorion or the placenta. All these specimens, although possibly wholly normal in development, might, if retained long enough, even under aseptic conditions, fall into the pathologic division. This division, as well as that for the normal embryos, no doubt also contains embryos and fetuses which have been affected by maternal uterine or constitutional diseases, especially by endometritis or lues. Since, as already stated, abnormal embryos or monsters are not grouped by themselves,
some of these also are included in the pathologic division, though most of them are classed as normal with localized anomalies. Moreover, it has been found that
"A certain number of specimens can not be classified satisfactorily in any of the groups, either normal or pathologic. For example, we have in our collection 4 specimens of cyclopia. One is in an embryo which is perfectly normal, except that it contains a single eye; the second is a stunted embryo which is more or less macerated; the third is a fetus compressus, and the fourth is a double monster. These have been entered in our catalogue as follows: Normal with cyclopia; stunted with cyclopia; fetus compressus with cyclopia; double monster with cyclopia. These varieties naturally include all specimens usually designated as monsters. The double monster mentioned above, curiously enough, becomes recorded as a normal twin embryo with cyclopia, but it is difficult to avoid this contradiction with the method of classification we have adopted."
The pathologic division, which does not comprise the greater number of specimens, is subdivided into seven classes, largely on the basis of the degree of destruction of the ovum or embryo. This holds unless we are to assume that the chorionic and amniotic vesicles can develop regardless of the presence of a cyemic primordium. Particularly in the first three groups, the degree of destruction of the cyema is the determining factor, for in all of the specimens in these groups it is missing. (1) The first group among the pathologic is composed of small remnants of specimens which usually are found to consist of villi only; most of these are derived from tubal pregnancies. (2) The second group is composed of small specimens which are represented by chorionic vesicles wholly devoid of contents save blood and the disintegration products of the embryo and amnion, blood, coagulum, or magma. (3) In the third group are placed specimens in which, in addition to the chorion, the amnion, in part or in its entirety, is present. These specimens usually are somewhat larger.
Although the absence of the cyema is the determining factor for the inclusion of a specimen in the first three groups in the pathologic division, its presence, no matter in how modified a form, is the criterion which places the specimen in one of the last four groups of this division. In general, it may be stated that the specimens in the last four groups are classified according to the degree of the change which they have undergone. "If the embryo is affected so as to appear as an amorphous nodule, it is placed in group 4; but if a cephalic or caudal extremity still can be recognized, it is placed in group 5." The word affected, as used here by .Mall, does not, however, carry with it the idea of dismemberment by force, for a macerated fragment of a normal embryo manifestly should not be classed here. However greatly the form of an embryo may have been modified, this modification must not have resulted from the mechanical removal of a portion of its substance. It is true that it is often impossible to determine whether a given modification in form has thus arisen. Sometimes, indeed, it is impossible to distinguish, by external examination alone, the stump of a cord or the yolk-sac from a nodular embryo, for in the main the classification of a specimen is determined by inspection alone with or without the aid of low magnification. Some of the speci-
46 STUDIES ON PATHOLOGIC OVA.
mens in the nodular group also contain rudiments of the extremities, for even with these present it has not always been possible to recognize the cephalic or caudal extremity by inspection alone. Since the position of the extremities ordinarily would suffice to make such recognition possible, it is necessary to recall in this connection how pronouncedly the body-form of the nodular embryos often is changed. In addition to such specimens, the nodular group contains three other kinds, such as stumps of the umbilical cord, isolated yolk-sacs, and sessile nodular embryos, many of which have nothing that could properly be regarded as a cord. In specimens classed in the fifth group of the pathologic division the embryo is destroyed almost entirely, but its head end can still be recognized. Hence it is designated as cylindrical, but the use of this term does not imply that these specimens are hollow, or wholly without limb-buds even.
"In this group are often included specimens of embryonic remnants which, upon close examination, prove to be the naked umbilical cord. As soon as such specimens are detected they are distinguished by a second entry cord. Unfortunately, these two types of specimens, although out of relation, fall together in our catalogue, but this discrepancy can not be avoided without giving up our plan of retaining the first entry of a specimen."
Indeed, specimens in which the body of the embryo has been so modified as to represent a rod are often indistinguishable upon inspection alone, not only from remnants of the cord, but also from rolled-up decidual or from cyemic remnants or polyps.
In the next or sixth group of the pathologic division, "the embryo can easily be recognized, but it is unusual in appearance, the head, extremities, or some other part of the body being atrophic, and for this reason this type is termed stunted. Usually all the parts are more or less degenerated. Most of these embryos are small, but they also can be classified according to their length. The histories show that these embryos are much older than normal ones of the same size, indicating that an arrest of development must have taken place." In these specimens the normal proportions also have changed. Moreover, since the adjective stunted implies a physiologic cause, it is necessary to add that in some of them the stunting, no doubt, is apparent only and is the result of distortion incident to, if not consequent upon, maceration. In such cases, then, the stunting is in no sense the consequence of an interference with the nutritive supply of the embryo.
Specimens of pathologic embryos and fetuses, mainly the latter, not included in the above six groups, are classed in group 7. "This last group is composed mainly of large specimens which are atrophied, dried up, or macerated. If the tissues are soft and edematous we call the specimen macerated; if it is dried up and atrophic we call it fetus compressus. In the course of time we may find that these two forms should be classed separately, but for the present it appears as if one is the forerunner of the other."
Anyone who reflects upon this classification will easily recognize its shortcomings. These were not overlooked by Mall. In the handling of a large series
of specimens, a simple, serviceable, even if rough classification is extremely useful. Moreover, under present circumstances, the first examination must be a gross and also a rather superficial one, unless technical assistance adequate to make a final diagnosis within a few days of the receipt of the specimen is available. This not having been the case, it has been found that a certain amount of shifting becomes inevitable unless the groups are to become mixed. The shifting necessitated by further examination is indicated in table 4. However, as the histories become
TABLE 4. Showing the shifting within the groups necessitated by microscopic examination.
No. of group.
No. originally in group.
Specimens transferred to
Final number in each group.
| 573 771o
1 80 53
f 21 123 147 770 791
130 469 143 531 180 658 257 803
279 960o 483 976 518 985 556 5846 689 701 7236 814
232 365 653 7716
(124 J649 1.797
25 433 32 446 37 441 166 528 189 610 198 674 692 244 795 253 807 275 830 291a 943 161 264 309 433a 573
142 414 246 419 347 399 150 205 289 400 655
188 601c 739 334 635e 802 348 65 la 962 375 681 983a 52 If 732 993
more complete, and the data which they contain also more reliable, and especially as our knowledge of chorionic and placental pathology becomes fuller, an entire reclassification may become advisable, if not inevitable. This, however, should not be cause for regret, for it is in the line of progress. Indeed, it was not unanticipated, and the revision indicated in table 4 was made at the request of Mall. It would seem, moreover, that a modification in classification is indicated also by the main thesis of Mall, that faulty implantation is the cause of abnormalities. Yet under the present classification, as Mall himself explained, cyemata with marked abnormalities frequently are classed as normal. Moreover, normal specimens are included among the pathologic division, which, strangely enough, contains no group of abnormals. This difficulty can be remedied easily, but how soon other difficulties indicated above can be overcome or be eliminated will depend very largely upon the further development and utilization of this already unique collection and upon the degree of cooperation which can be secured from clinicians and administrators.
After considerable reflection upon this classification, and some experience in its daily application to material as received from physicians, I shall permit myself certain suggestions, especially since they were invited by Mall himself. It would seem that a division of all conceptuses into the long-established classes of normal and abnormal has certain advantages. This would be true even if it were as difficult to discriminate between the normal and the abnormal or atypical as between the normal and the pathologic. A careful structural examination can always decide the former, at least after the establishment of an arbitrary limit for the range of normal variation. But until we can recognize the earliest morphologic and cytologic changes resulting from pathologic conditions, it often must remain impossible to decide between the normal and the pathologic. That day seems to lie far in the future much farther, indeed, than the possibility of a final determination regarding the normality or abnormality in form of early cyemeta or chorions.
A division into normal and abnormal also would avoid the awkward necessity of classifying monsters as normal and of putting all specimens composed of normal villi, of normal empty, damaged, chorionic vesicles, and of normal ruptured chorionic and amnionic vesicles into the pathologic division. A grouping into the normal and abnormal also would enable us to frankly recognize the fact that normal specimens may become decidedly macerated, but should therefore not be grouped among the pathologic. Furthermore, it would enable us to recognize the possibility, first emphasized by Giacomini, that normal cyemeta possibly may be changed so as to perfectly simulate the atrophic and stunted forms.
The division of abnormals also could be understood as including not only specimens resulting from anomalous development, i. e., true malformations, but also those which have been affected by disease. Some of the latter, if not anomalous also, would fall in the group of normals, as they rightly should. Moreover, unless it can be confidently assumed that abnormal ova and spermatozoa have no existence, some of the cyemata now classed as pathologic evidently are not the direct products of disease. A specimen certainly may be abnormal without being pathologic and pathologic without being abnormal in form.
Phisalix (1890) also called attention to the fact that an embryo which is pathologic is not necessarily abnormal. A fetus that dies in utero while the mother is suffering from an infectious disease may well be, and often is, pathologic, but it is not necessarily abnormal. Neither is a specimen necessarily pathologic which dies under fortuitous circumstances and is retained sufficiently long so that maceration changes very materially alter its form. Likewise, fetuses which die as a result of extreme twisting or strangulation of the cord, or from premature separation of the placenta, or from some other accident of pregnancy, are not necessarily pathologic. Yet all these, if retained sufficiently long, might, under the present classification, fall into this division.
Although we know very little about the existence of abnormal ova, the relative frequency with which anomalous forms of spermatozoa occur would alone seem to make decidedly venturesome the assumption that there is no such thing as germinal anomalies. In view of the fact that we frankly recognize the occurrence of hereditary anomalies, germinal causes certainly can not be excluded. Yet I presume it would be difficult to refer to such specimens as pathologic unless we extended the customary meaning of this term very materially.
After grouping all specimens among the abnormal and the normal, each of these classes could then be subdivided very much as the group of pathologic specimens under the Mall classification i. e., fragments of villi or vesicles may be either normal or abnormal; and I think it will be possible to show that the same may be true, certainly of the macerated, and probably also of most of the stunted cyemeta, for death in these may have been due purely to mechanical or other causes, which have not resulted in any abnormality of development, the changes in external form being the result of postmortem antepartum changes, as suggested by Giacomini.
By the addition of three groups before that composed of villi only, all abortuses would seem to be provided for. These groups would be composed of those containing syncytial remnants only, those containing trophoblast only, and those containing both syncytium and trophoblast. However, since as Muller (1847) emphasized, normal embryos may be found in diseased vesicles and abnormal embryos in normal vesicles, it also will be necessary to classify vesicles and cyemata separately whenever both do not fall into the same group.