AN 
EXPERIMENTAL STUDY 
OF 
GLOMERULONEPHRITIS. 
BY 
WILLIAM H. WELCH, M.D., 
BALTIMORE. 
TRANSACTIONS OF THE ASSOCIATION OF AMERICAN PHYSICIANS, 
REPRINTED FROM THE 
June 18, 1886. 
WM. J. DORNAN, PRINTER. 
1886. 
PHILADELPHIA: AN 
EXPERIMENTAL STUDY 
OF 
GLOMERULONEPHRITIS. 
WILLIAM H. WELCH, M.D., 
BY 
BALTIMORE. 
REPRINTED FROM THE 
TRANSACTIONS OF THE ASSOCIATION OF AMERICAN PHYSICIANS, 
june 18, 1886. 
PHILADELPHIA 
WM. J. DORNAN, PRINTER. 
1886. AN EXPERIMENTAL STUDY OF GLOMERULO- 
NEPHRITIS. 
WILLIAM H. WELCH, M.D., 
BALTIMORE. 
Of the various processes which make up the pathological anatomy 
of Bright’s disease, perhaps the two which at present awaken the 
greatest interest and the study of which promises the most fruitful 
results, are the changes which take place in the glomeruli and 
atrophy and necrosis of the epithelial cells in relation to interstitial 
changes. Not that by any means unanimity of opinion has been 
reached regarding other fundamental questions involving the relation- 
ship between the glandular, the interstitial, and the vascular lesions of 
Bright’s disease, but the two processes named have acquired especial 
prominence within the last few years by the recognition of their fre- 
quency and importance. 
Alterations of the Malpighian bodies, with which alone the present 
paper is concerned, were described with much accuracy and fulness 
by Beer in his excellent treatise on The Connective Tissue of the 
Human Kidney in the Healthy and the Diseased State, published 
in 1859, but until recently little attention was paid to his observa- 
tions. Klebs, by his description in 1876 of the lesion to which he 
gave the name of glomerulo-nephritis, incited renewed investigation 
of this subject. During the last ten years processes embraced under 
the name of glomerulo-nephritis have been studied by Cornil, Hortoles, 
Friedlander, Bibbert, Nauwerck, and many others, but we owe to 
Langhans the most extensive and accurate observations on this 
subject. 
A comparison of the statements of different writers concerning 
glomerulo-nephritis will show a wide diversity of views, not only as 
to the interpretation of the facts observed but as to the facts them- 
selves. Without attempting a critical analysis of these divergent 4 
WELCH, 
views, the following summary of some of the conclusions of different 
observers will show this to be true. 
At present little or no credence is given to Klebs’s belief that the 
essential lesion of glomerulo-nephritis is proliferation of connective 
tissue cells which he supposed to exist between the capillaries of the 
glomerulus, although this belief still receives some support from the 
conception of Hortoles, and of Cornil and Brault, as to the connec- 
tive-tissue nature of these intercapillary cells. 
Langhans, in his publication in 1879, dwelt especially upon swell- 
ing, proliferation, and desquamation of the epithelial cells both of 
the glomerulus and of Bowman’s capsule; whereas in his last article, 
published in 1885, he regards swelling and proliferation of the endo- 
thelium lining the glomerular capillaries as the primary and essential 
lesion of glomerulo-nephritis, as indeed of all forms of acute nephritis. 
Similar views are held by Friedlander and by Nauwerck. 
Ribbert considers that the only changes are swelling and des- 
quamation of the glomerular and capsular epithelium, changes which 
he regards as essential to all forms of acute and chronic Bright’s 
disease and as answerable for many of the symptoms. He, as well 
as Hortoles and Cornil, refers the appearances described by Langhans 
as proliferation of the capillary endothelium to an accumulation of 
white blood-corpuscles in the capillaries. 
Cornil and Brault assign an insignificant and inconstant role to 
changes in the epithelium of the glomerulus and of its capsule, and 
find the fundamental lesion of acute glomerulo-nephritis to be an 
exudation from the glomerular capillaries of inflammatory products 
composed of white blood-corpuscles, red blood-corpuscles, and an 
albuminous fluid. 
On the other hand, Hortoles and Ribbert deny altogether the 
occurrence of emigration through the glomerular capillaries, basing 
this view on the failure to find leucocytes in the capsular space 
when their presence cannot be explained by emigration from capil- 
laries adjacent to Bowman’s capsule. In further support of their view 
they urge the peculiar structure of the capillaries of the glomerulus. 
But it is not necessary to consume time in a further enumeration 
of the different views which are held upon this subject. Enough 
have been cited to show that there is room for much more investiga- STUDY OF GLOMERULO-NEPHRITIS. 
5 
tion, as well as to bring out some of the leading points which require 
further study. 
In the expectation that light might be thrown upon some of these 
doubtful points by experiments upon animals, I have made a study of 
the nephritis produced by acute cantharidin poisoning with especial 
reference to the alterations induced in the Malpighian bodies. I was 
led to select cantharidin because several experimenters with this 
substance have described in cantharidin nephritis notable changes in 
the Malpighian bodies, and because Cornil bases his description of 
acute glomerulo-nephritis chiefly upon observations of the kidneys of 
rabbits poisoned by cantharidin. 
My experiments, which have not yet reached their completion, 
were made upon white rats and upon rabbits. A concentrated solution 
of cantharidin in acetic ether was used. I injected, subcutaneously, in 
rabbits usually from one-half to one centigramme of cantharidin, and 
in rats from one to three milligrammes. When a number of injec- 
tions were made upon successive days smaller doses w7ere used. The 
number of injections did not exceed five, and usually not more than 
two or three. 
After the injection of a toxic dose the urine becomes diminished 
and finally suppressed; it contains albumen, hyaline casts, and a 
large number of leucocytes and red blood-corpuscles. The kidneys 
appear swollen, congested, and more succulent than normal. 
The microscopical appearances in the rat’s kidney will be first 
described. Here and there foci of infiltration with small round cells, 
doubtless emigrated white blood-corpuscles, can be found. They are 
most common around the veins at the base of the pyramid. They 
are often hut not constantly present. 
The epithelium of the convoluted tubes is in places normal in 
appearance, in other places it is swollen, and often the inner part of 
the epithelial cells is broken off and appears as a granular mass in 
the lumen of the tubes. Sometimes a large number of epithelial 
cells are devoid of nuclei, and have apparently undergone coagulation 
necrosis. In general, the convoluted tubes are wider than normal. 
The most marked changes are to be found in the Malpighian 
bodies. In by far the greater number of these bodies there is 
between the glomerulus and Bowman’s capsule a wide space partly or 
wholly filled with cells and granular material. These cells may be 6 
WELCH, 
round or oval, but they are usually polyhedrical in shape. They are 
much larger than white blood-corpuscles, havft granular protoplasm, 
and oval, vesicular nuclei. They are frequently arranged as a cres- 
centic mass around the glomerulus. 
Evidently we have here appearances such as have been often 
described as glomerulo-nephritis in the human kidney, and naturally 
one seeks for the same origin of these cells as that assigned for the 
cells similarly situated in human glomerulo-nephritis, namely, swelling 
and desquamation, and possibly proliferation of either the capsular 
or the glomerular epithelium. 
The microscopical appearances in the rat’s kidney, however, do not 
admit of this explanation. The epithelium lining the capsule of 
Bowman is often intact, and can be traced in its normal situation and 
with its normal appearances around the mass of cells occupying the 
capsular space. The glomerular epithelium may also preserve its 
normal position; more frequently it is somewhat swollen and granular, 
and it may desquamate. Such desquamated glomerular epithelial 
cells may be mingled with the cells free in the capsular space, but 
there are no appearances which justify the derivation of the majority 
of these latter cells from the epithelium of the glomerulus. 
Both Cornil and Eliaschoff, whose descriptions of cantharidin 
nephritis are the best which I have found, and who experimented on 
rabbits, also came to the conclusion that the cells occupying the cap- 
sular space cannot be derived from either the epithelium of Bowman’s 
capsule or that of the glomerulus. They argue that there remains 
hut one other possible source, namely, emigration of white blood- 
corpuscles from the glomerular capillaries. They therefore consider 
the cells in question as white blood-corpuscles, of which the cell 
bodies and the nuclei are greatly swojjen and altered by the action of 
the cantharidin or of the urine. 
So far as the rat’s kidney is concerned, this explanation of Cornil 
and of Eliaschoff cannot he admitted. These authors are in error in 
supposing that the cells occupying the capsular space, if not originat- 
ing from the capsular or glomerular epithelium, must come from the 
capillaries. These cells may be derived also from the convoluted 
tubes in immediate communication with the Malpighian bodies, and 
that this is their origin, at least in part, in the cantharidin nephritis 
of rats can be proven, I believe, beyond question. The cells in the Cantharadin Nephritis . White Rat. 
A. Iloen ft Co. Lifli. Bultinuire. STUDY OF GLOMERULO-NEPHRITIS. 
capsular space are identical in appearance with the epithelial cells of 
the adjacent convoluted tubes. The appearances presented on sec- 
tions which show the communication between the capsular space and 
the corresponding convoluted tube hardly admit of any other inter- 
pretation than that which I have given. Here it can be seen that 
the mass of cells in the capsular space is in direct connection with 
the epithelium of the convoluted tube, and, what is especially demon- 
strative, there can frequently be found in this mass a group of cells 
arranged regularly in the form of a ring with a central space, just 
like the epithelium of a uriniferous tube. Without the explanation 
given, such an arrangement of the cells is, of course, very puzzling, 
especially when the Malpighian body is cut so as not to show the 
connection with the tube. The appearance is as if the epithelial cells 
lining the mouth of the uriniferous tube (or the neck of the capsule), 
together with the cells immediately adjacent, had been pushed 
mechanically into the capsular space. A similarity in appearance 
between the cells accumulated in the capsular space and the epithelial 
cells of the convoluted tubes, might be explained by the fact that 
under normal conditions the epithelium of the convoluted tubes may 
extend for a variable distance along Bowman’s capsule. Such an 
extension of the tubal epithelium to the capsule is not particularly 
noticeable in the rat’s kidney, and cannot be adduced to explain the 
peculiar appearances described. An inspection of a microscopical 
section or of a drawing will, I believe, leave no doubt as to the cor- 
rectness of the interpretation which I have given*. (See Plate.) 
Observations concerning the passage of epithelial ceils from the 
convoluted tubes into the capsular space, have been made by Kol- 
liker1 and by Argutinski.2 While discussing the nature of the 
normal glomerular epithelium, Kolliker says : 
“ In hardened kidneys the epithelium of the convoluted tubes is not infre- 
quently pressed up into the capsules, so that in many cases a membranous 
layer is formed which is arranged like a funnel around one end of the 
glomerulus.” 
Kolliker’s observations relate, therefore, to artefacts produced by 
the action of the hardening fluid. In the cases of cantharidin 
1 Kijlliker, Handb. d. Gewebelehre, Leipzig, 1867, p. 504. 
2 Argutinski, Beitrage zur Normalen und Patkologischen Histologie der Niere, p. 18. Inaug. Diss. 
Halle, 1877. 8 
WELCH, 
nephritis which I have studied, the accumulation of tubal epithelium 
within the capsules cannot be regarded as an artefact, for the same 
appearances can be seen in sections of the fresh kidney as in sections 
of kidneys hardened in Miiller’s fluid, chromic acid, Flemming’s 
solution, picric acid, osmic acid, and alcohol. The change is, there- 
fore, one which occurs during the life of the animal. The same 
appearances in the Malpighian bodies which I have observed in the 
cantharidin nephritis of white rats, have been described and figured 
by Argutinski as occurring in the embolic infarctions experimentally 
produced in dogs by injecting into the renal arteries plugs of wax. 
He also speaks of the possibility of producing the same change by 
forcible injection of fluids into the renal bloodvessels. He therefore 
assigns as the cause of this phenomenon the pressure exerted by 
dilated bloodvessels upon the convoluted tubes, whereby the proto- 
plasmatic contents of the latter are forced upward into the capsular 
spaces. 
The following are among the factors which may be adduced to 
explain the mechanism by which the cells in cantharidin nephritis 
make their way from the convoluted tubes into the capsular spaces: 
Obstruction of the uriniferous tubes, both in the cortex and in the 
pyramid, by desquamated and by necrotic epithelial cells, and by 
hemorrhages within the tubes, pressure upon the tubes from without 
by congested bloodvessels and by transuded serum, and cessation of 
the excretion of urine from the glomeruli. In consideration of these 
alterations in the kidney, it is comprehensible that swollen and 
desquamated epithelial cells in the beginning of a convoluted tube 
might find toward the capsular space the direction of least resistance. 
There is, of course, no propriety in designating as glomerulo- 
nephritis the changes in the Malpighian bodies of the rat’s kidney 
which have been described. The term glomerulo-nephritis, however, 
is one which is used with much latitude of significance, and embraces 
nearly all of the changes observed in acute nephritis in the Mal- 
pighian bodies and a large number of those found in chronic nephritis. 
Many of these changes are not proven to be inflammatory. 
I am not able to say whether a process similar to that which I 
have found in the Malpighian bodies of the kidney of the rat poisoned 
by cantharidin, occurs also in the nephritis of other animals and of 
man. In the acute cantharidin nephritis of rabbits the capsular spaces STUDY OF GLOMERULO-NEPHRITIS. 
9 
also contain granular material and cells, but the cells do not form such 
compact masses as in the rat’s kidney, and more frequently undergo 
necrosis. I have not been able as yet to reach a positive conclusion 
as to the origin of these cells. To the descriptions which have been 
repeatedly given of the cantharidin nephritis of rabbits, I have to 
add the almost constant occurrence of necrosis of epithelial cells in 
certain tubes occupying the boundary zone of the pyramid and the 
medullary rays. These tubes appear to be the terminal portions of 
the proximal convoluted tubes (Endstiickchen, spiral tubes). Hem- 
orrhages within the tubes, particularly those of the pyramid, are not 
infrequent. 
In human glomerulo-nephritis one is not generally at a loss to account 
for the crescentic mass of cells which frequently occupies the capsular 
space. The presence of these cells is accompanied with swelling and 
desquamation of either the glomerular or the capsular epithelium, 
usually of both, and these changes explain the accumulation of cells 
in the capsular space. 
One of the leading objects in making these experiments on cantha- 
ridin nephritis was to determine whether changes were thereby pro- 
duced in the glomerular capillaries. Friedliinder, Nauwerck, and 
Langhans have called attention to the accumulation of cells in the 
interior of the glomerular capillaries as a frequent and important 
lesion of acute and of chronic nephritis. These cells they regard, for 
the most part, as proliferated capillary endothelial cells. Previous 
experimenters on nephritis make no mention of alterations in the 
glomerular capillaries, so far as I have been able to learn. 
The technical difficulties attending the microscopical investigation 
of the glomerular capillaries are considerable. It is often desirable, 
as recommended by Langhans, that these capillaries should be in- 
jected with colorless gelatine, although when they are widely dis- 
tended with cells, as in some instances of Bright’s disease, such a 
procedure is not necessary. The sections of the glomeruli must be 
extremely thin, and this is best accomplished by embedding pieces of 
the kidney either in celloidin or in paraffin, in the latter case cutting 
the sections, of course, with the dry razor placed at right angles to 
the specimen. 
In some cases of cantharidin nephritis no changes can be detected 
in the glomerular capillaries ; in some instances, however, especially in 10 
WELCH, 
the rabbit, these capillaries contain a large number of small, darkly 
staining nuclei, readily distinguished from the larger and paler nuclei 
of the glomerular epithelium. These nuclei belong for the most part 
to spherical, granular cells, which I am inclined to interpret as white 
blood-corpuscles. Positive evidences of proliferation of the capillary 
endothelium could not be found. Changes in the capillaries compar- 
able in degree to those observed in many cases of nephritis in human 
beings, I have not been able to find in experimental cantharidin 
nephritis. 
The occlusion of the glomerular capillaries in acute and chronic 
Bright’s disease by granular material and cells, has been most fully 
described in all of its histological details by Langhans. My own 
studies of human kidneys are essentially confirmatory of Langhans’s 
descriptions. 
Swelling of the endothelium, and accumulation of cells in the 
glomerular capillaries, are nearly constant lesions in acute scarlatinal 
nephritis. These changes in the capillaries may be the only marked 
lesions in the kidney, although, as a rule, there is more or less accumu- 
lation of lymphoid cells in the interstitial tissue. I have observed 
similar accumulation of cells in the glomerular capillaries in acute 
Bright’s disease complicating typhoid fever, and in a number of cases 
of chronic Bright’s disease. The changes in the capillaries are usually 
accompanied by other lesions of diffuse nephritis. 
I would, however, call especial attention to certain cases of Bright’s 
disease which usually pursue a subacute course, in wThich the accumu- 
lation of cells in the capillaries of the glomerulus is the predominant, 
and sometimes the only evident lesion. The following case may serve 
as an illustration. 
The patient, who had had malaria, was the subject for two months 
of anasarca. The urine was scanty, albuminous, and contained casts. 
Death resulted from uraemic convulsions. At the autopsy there was 
found malarial pigmentation of the spleen and other organs, the heart 
was considerably hypertrophied, and the kidneys were large, the 
surface smooth, the capsule non-adherent, and the cortex swollen. 
The Malpighian bodies were large and pale. Upon microscopical 
examination of the kidneys, with the exception of the change about 
to be described, there was very little abnormal. The epithelial cells 
were normal. Only after careful search could a few patches of infiltra- STUDY OF GLOMERULO-NEPHRITIS. 
tion with lymphoid cells be discovered. The Malpighian bodies were 
rich in nuclei. The glomeruli filled completely the space enclosed by 
Bowman’s capsule. The case seemed a puzzling one, until after the 
examination of very thin sections it was found that almost everywhere 
the glomerular capillaries were dilated, and contained a large number 
of cells, partly resembling white blood-corpuscles, but mostly larger, 
and of an endothelioid type. 
Probably many pathologists have met with kidneys in which the 
lesions seemed entirely inadequate to explain the symptoms. Doubt- 
less a certain number of these obscure cases belong to the class just 
described. It certainly is important that a careful examination be 
made of the condition of the glomerular capillaries. 
It is not necessary to dwell upon the importance which attaches to 
the lesions of the glomerular capillaries which have been described. 
When one considers the physiological functions of the glomeruli, it is 
difficult to think of any lesion of the kidney more destructive of its 
functions than the occlusion of the capillaries of the glomerulus. It 
is more reasonable to refer the production of albuminuria to changes 
in the capillary walls of the glomeruli than to desquamation or other 
alteration of the glomerular epithelium. Certainly some swelling and 
desquamation of this epithelium are extremely common, even in cases 
without renal symptoms. 
But while I am disposed to attach such importance to alterations 
of the glomerular capillaries, we are not justified in asserting, as has 
been done, that these changes constitute the primary and essential 
lesion of all cases of Bright’s disease. The glomerular lesions are 
coordinate with the parenchymatous, the interstitial, and the other 
vascular lesions of diffuse nephritis. 
As regards nomenclature, there is not much use in fighting against 
names which have gained currency, although the term glomerulitis 
seems to be more suitable than glomerulo-nephritis, and equally expres- 
sive. It is customary to include all of the changes of the Malpighian 
bodies which are not either atrophic or purely degenerative, under the 
name glomerulo-nephritis; but it is to be borne in mind that this term 
is used to designate a group of lesions belonging to diffuse nephritis, 
and not a disease by itself. I would suggest that the form character- 
ized by an accumulation of cells between the glomerulus and Bow- 
man’s capsule be designated desquamative glomerulitis, and that characterised by accumulation of cells, or other changes in the interior 
of the capillaries, be called intra-capillary glomerulitis, and this with- 
out prejudging the question as to the propriety of considering all of 
these changes as inflammatory. 
WELCH, STUDY OF GL O ME EULO - N EP HR ITIS.