Frederick M. (Frederick Madison) Allen.

Total dietary regulation in the treatment of diabetes online

. (page 67 of 76)
Online LibraryFrederick M. (Frederick Madison) AllenTotal dietary regulation in the treatment of diabetes → online text (page 67 of 76)
Font size
QR-code for this ebook

the power of the pathologist. In the main, however, attention has
been centered upon the grosser structural disorganization, and the
carelessness of pathologists regarding finer cellular changes is respon-
sible for their overlooking the vacuolation accurately described and
pictured by Weichselbaum and Stangl. It is furthermore important,
in estimating dearth of islands, to recognize th* pseudo-islands which
occur in some cases, probably from proliferation of duct cells, and
which lack both the structure and function of true islands.

Third, the clinical grouping of cases has been misleading. Inher-
ently mild diabetes has been classed as severe merely because of
acetonuria or perhaps an unnecessary death in coma. Maximal
severity has been attributed to certain cases on the ground of their
rapid course and early fatal termination. Since such patients when
given the opportunity by proper treatment often manifest a surpris-
ingly high assimilative power, the finding of numerous islands in their
pancreas is now seen to be in strict accord with the insular theory,
instead of inimical to it as formerly supposed. With a truer clinical
classification, and with treatment preventing a large proportion of
early unnecessary deaths, pathologists will find the inicroscopic diag-
nosis much simpler. On the other hand, the clinical statement that
diabetes was absent need not be accepted as infallible. Generally it
means only absence of glycosuria. Tests of the blood sugar and glu-
cose tolerance have seldom been made, and even these may be viti-
ated by cachexia or other disturbances. Weichselbaima records one
instance in which the pathologic findings pointed to diabetes, which
was not clinically evident during the final iUness in the hospital,
but careful inquiry of the relatives elicited a history of typical diabetic
symptoms previously. In most diseases the pathologist corrects the
clinician, not vice versa; and this will be increasingly true of diabetes
as the morbid anatomy is better understood.


The diagnostic import of the island exhaustion resulting from dia-
betes, and of the lesions causing diabetes, may be summarized as

1 . Hydropic Degeneration. — It is necessary to define the signifi-
cance of the presence and absence of this change.

Its Presence. — Vacuolation of cells is an appearance which patholo-
gists are likely to view with suspicion as an accidental or unreHable
phenomenon. Forttmately the islands of Langerhans are little sub-
ject to non-specific vacuolation. Fry'^ mentions vacuolation and
degeneration of the islands in a case of myxedema. The bare possi-
bility of masked diabetes must be considered; or, as vacuolation of
the adrenals is similarly mentioned, it is conceivable that the change
in the islands may have been fatty. Extensive examination of mis-
cellaneous autopsy specimens has thus far confirmed Weichselbaum's
conclusion in favor of the specificity of the true hydropic alteration.
Careful study is still needed to decide whether actual fluid accumu-
lation ever takes place in the island cells from any cause other than
diabetes, and whether fatty or other changes ever simulate the real
exhaustion. The most frequent cause of uncertainty is imperfect
preservation of tissue. Shrinkage, as from formaldehyde, small
breaks in the sections, and other artefacts may be confusing. Especi-
ally, with only a little delay in fixation, the dehcate menlbranes of
exhausted cells may break down first of all; and with later autopsies
blurring and fragmentation of the islands may make pictures hard
to interpret. Under these circumstances simple pallor of island cells,
and the finding of naked nuclei, is suggestive but not conclusive.
With abundant islands present, there may then be justified hesita-
tion between two extreme judgments; one, complete absence of dia-
betes, with perhaps nothing wrong but a trivial pancreatitis; the
other, diabetes of the most intense rapid ty^Q with early death in
coma. Diagnosis is sometimes demanded not only from poorly
fixed tissue but also from single sUdes, whereas search through all
portions of the pancreas is required by the fact that certain areas
sometimes reveal pictures not seen in the others. It seems safe to
say that the typical picture of vacuolation in some cells and pyknotic

35 Fry, H. J. B., Quart. J. Med., 1914-15, viii, 276-299.


nuclei and degeneration in others is absolutely diagnostic of diabetes.
Furthermore, if in satisfactorily prepared tissue an observer familiar
with this process finds a single island cell which is unmistakably
swollen and vacuolated, with a nucleus (generally normal) surrounded
by a halo of clarified cytoplasm inclosed within a distinct cell mem-
brane, in sharp contrast to the other normal appearing island cells,
the existence of diabetes is estabhshed. Only, the warning against
accidental appearances more or less closely imitating the true pic-
ture must here be emphasized. As the sign is one of active diabetes,
the clinical record regularly agrees; but if any exception is found,
the cHnical diagnosis should be subjected to fully as close scrutiny
as the pathologic diagnosis.

Its Absence. — In animals the hydropic change is a fairly reUable
and delicate index of diabetes, beginning with slight vacuolation of a
few cells in the early mild stage, and advancing parallel with the
duration and severity. It appeared puzzling that a considerable pro-
portion of human cases, surpassing in both these respects many of
the animals, show entire absence of visible exhaustion; yet the ex-
planation is simple. In the strictest sense, the hydropic degeneration
is an expression not of diabetes but of cellular dissolution due to
diabetes. It is unknown at what point repair becomes impossible;
but in animal experiments a cell evidently persists only a few days
in its terminal stage of extreme vacuolation and swelling. Under
sharp dietary restriction, gradual recovery seems to be possible even
at this stage, apparently corresponding to the marked transformation
and gain of tolerance in some human cases on fasting. Otherwise,
cells and islands break down and are permanently lost, as already
explained. Human cases seldom progress as rapidly as the experi-
mental diabetes of animals, in which the destruction of islands may
be practically complete within a few weeks or months, as proved
anatomically and by the inability to become sugar-free on fasting.
The human cases are more likely to last months and years, and even
in the worst forms to retain enough island tissue to enable control
by fasting. Accordingly, the visible degeneration could not be so
striking in the human pancreas as in animals unless the clinical prog-
ress were as rapid. Moreover, the intensity of clinical symptoms
is not a sole criterion of the anatomic effects. Not only may a dog

640 CHAPTER vin

with a 2.8 or lower D: N quotient show more rapid and extensive
island destruction than a man with a 3.65 :1 ratio, but the differences
between human patients are still greater. It is well known that
cases with similar hyperglycemia, glycosuria, and acidosis differ
widely in progressiveness, such differences being most familiar be-
tween elderly and youthful patients, but sometimes equally prom-
inent between different cases at the same age. Specific differences
in the susceptibihty of the island cells to anatomic breakdown
from functional overstrain must be recognized. Therefore it is
impossible to establish any rule as to the proportion of cases in
which hydropic changes wiU be foimd. It can only be said that
they, are invariably present when the intensity and the duration of
the diabetes are sufl&cient. They are practically always present in
typical coma, though when this foUows only a short period of intense
symptoms, they may sometimes be surprisingly slight and hard to
demonstrate. Specimens from severe cases of diabetes treated by
the old methods show them in the majority of instances. They are
less frequent in proportion as the diabetes is milder or the treatment
more efficient. The previous statement that persistent hypergly-
cemia without glycosuria causes degeneration of islands is a deduction
as applied to human cases. The evidence for it is that diabetic ani-
mals under these circinnstances show gradual loss of tolerance, and
slight hydropic changes are demonstrable in their islands. Human
patients show a similar injury of assimilation, which is sometimes so
slow that it may appear doubtful but is nevertheless real; but the
anatomic process has not as yet been observed in them under these
conditions. Such a gradual loss, probably no more than the disin-
tegration of an occasional scattered cell, though real, will necessarily
be seldom convincingly visible. Diminished frequency of hydropic
degeneration should be the accompaniment and proof of improved

In brief, the presence of hydropic degeneration may be held to
establish a positive diagnosis of diabetes, but its absence is not
proof that diabetes is absent.

2. Lesions Causing Diabetes. — ^AU pathologic researches on the
subject have agreed concerning the occurrence of pancreatitis, and
notably of fibrous, hyaline, fatty, pigmentary, malignant, or other


changes affecting the islands, in a variety of conditions without posi-
tive diagnostic significance, but on the whole more frequently and ex-
tensively in diabetic than in non-diabetic cases. In the nature of
things, it has been impossible to establish any uniform rule for the
diagnosis of diabetes from such observations, for several obvious
reasons. First, valid judgment of the extent of island destruction
requires unusually painstaking study, because of the differences even
between closely adjacent areas of the normal pancreas, as shown by
Bensley, and still more so in pathologic material. Second, some
allowances must be made for functional influences and differences, as
mentioned previously. Third, the pathologist cannot be expected to
estimate concomitant clinical conditions. The characteristic active
symptoms of diabetes may be absent with emaciation from anorexia,
impaired food absorption, or any form of cachexia, when the pancre-
atic destruction is such as would certainly bring on these symptoms
in a patient otherwise healthy. The condition is entirely similar
to the apparent absence of diabetes in cachectic dogs possessing only
insignificant atrophic remnants of pancreas. This is doubtless the
chief explanation of the usual lack of diabetic symptoms with pan-
creatic cancer. In the main, the inabiUty to draw a sharp pathologic
line between diabetes and non-diabetes merely expresses the fact
that no such hne exists. There are all gradations between normal
and diabetic, with or without demonstrable impairment of food
assimilation. With a certain allowance for functional variations, any
person or animal is diabetic in proportion as islands of Langerhans
are lacking, and this anatomic fact is so thoroughly established that
it is independent of clinical complexities. Whereas the uncertainty
of diagnosis might formerly be imputed solely to the ignorance of
the pathologist, the latter should now be able to recognize diabetes
positively in any case of fatal severity, without regard for func-
tional differences, from the anatomic deficiency of island tissue
alone, and to form also some idea of the degree of true severity.
When islands are so abundant as to make the diagnosis on this basis
doubtful, the pathologist may say positively that, barring disobedi-
ence, complications, or other uncontrollable accidents, death was due
to inadequate treatment and not to actual severity of the diabetes.

642 CHAPTER vxn

Several trials have been made of the diagnosis of unlabelled pan-
creas specimens from various sources. In brief, when the fixation
has been good and the diabetes severe, the distinction from miscel-
laneous non-diabetic material, on the combined basis of vacuolation
and scarcity or "atrophy" of islands, has succeeded in almost 100
per cent of cases. When either the fixation was poor or the diabetes
mild, mistakes have been so numerous that the results were
largely unreliable.

The confusion concerning pancreatic pathology in the past has
wrongfully cast doubts upon the pancreatic origin of diabetes, not-
withstanding the wide discrepancies between structure and function
freely recognized for other viscera. More accurate correlation of all
the facts, clinical, experimental, and anatomic, justifies the inquiry
with what other organ or disease is the relation between function
and structure so clear and demonstrable as in the case of the pancreas
and diabetes.

C. Relation to Treatment.

For their therapeutic significance, the anatomic changes are again
conveniently divisible into those caused by and causing the diabetes.

1. Hydropic Degeneration. — Clinically, three lessons stand out.
One is the establishment of diabetes as an independent condition with
a pathology of its own. On the one hand, this contributes further to
clear away the pernicious confusion of diabetes with non-diabetic
forms of clinical and experimental glycosuria. On the other hand,
diabetes is raised above the rank of a mere symptom of pancreatic
disease. Its right to recognition as a distinct morbid entity has long
been generally acknowledged because of its importance and its broad
chnical characteristics. It is now seen to have a pathologic course
independent of the inflammatory or other basis on which it arose.
Treatment of diabetes by diet therefore is not mere palliation of a
symptom, but is a genuine means of checking the progress of a
definite pathologic change.

Second, anatomic guidance is given for the direction, ideals,
and limitations of dietotherapy. It emphasizes the need of beginning
at the earhest possible stage, when islands are most abundant and
the power of regeneration perhaps also exists. The fact should be


emphasized that, whatever excuses may be made for lax treatment,
it is responsible for progressive destruction of islands of Langer-
hans. With grossly excessive diet and flagrant symptoms the loss
is rapid; with only hyperglycemia and perhaps traces of acidosis
it is slower; also the susceptibility to functional overstrain differs
widely in different cases; but the ultimate injury manifests itself
clinically, without exceptions other than a few cases so mild that
they tend strongly to recover. Frequently the patient comes
under thorough treatment only after his own carelessness or a physi-
cian's mistakes have caused irreparable downward progress. In the
stage of greatest severity, exemphfied by most of the cases of the pres-
ent series, the remaining island tissue is very scanty, and any ca-
pacity of regeneration seems to be exhausted. The inability to re-
cover assimilative power, and the difficulties in any research aiming
at a cure of such cases, are made clear by this depletion not only of
islands but apparently also of the cells in ducts or elsewhere which are
able to form islands. The purpose of diet is therefore to prevent the
specific degeneration of the islands by relieving functional overstrain.
It obviously cannot check destruction by progressive inflammatory
or other processes. There is also a bare possibility that Weichsel-
baum's "atrophy" may represent a hopelessly progressive disease
of the islands. Experience shows, however, that the signs of pro-
gressive inflaramatory change are usually greatest in the more elderly
patients, whose diabetes is as a rule milder and easier to control,
while in a large proportion of youthful patients the progressive loss
of islands is due chiefly to hydropic degeneration. As mentioned,
the animal experiments demonstrate the possibility of preserving the
islands and the tolerance by restriction of all classes of food in the
absence of any progressive tendency. Though the clinical results are
marred by faults of patients and by mistakes in the apphcation of
the treatment, they prove that in at least a large proportion of cases
an existing downward progress can be checked and a stationary or
improved assimilation maintained through months or years. To date,
as far as the principle of total dietary regulation has been faithfully
carried out, only one case has yet been observed of apparent
"spontaneous downward progress" in absence of the metaboHc injury
of improper diet or infectious complications. Therefore, whatever


part may be assigned to the causes back of the diabetes, it must be
concluded that degeneration of islands due to errors in dietetic man-
agement has been a leading factor in the supposedly "spontaneous"
aggravation of the condition.

Third, a verdict on the treatment can in large measure be given
from the autopsy. Weichselbaum and Stangl's cases were doubtless
treated by the former orthodox method, and the hydropic degenera-
tion found in a majority of them is positive proof of rapid destruction
of islands due directly to the diet. Allowance must be made for dis-
obedience, intercurrent infections, and other uncontrollable accidents.
Otherwise, the finding of hydropic changes condemns the treatment.
Vacuolated islands are not present with thorough dietary control.
It seems possible also to form a correct anatomic judgment of the
true severity of the diabetes. With close distinctions between prep-
ositions, it may be said that many patients die with or from, but
very few of diabetes. They die with diabetes if they are carried off
by some extraneous cause. They may die from diabetes indirectly, if
death results, for example, from infection for which diabetes is re-
sponsible through lowering resistance; or more directly, if the diabetes
reaches such intensity that fasting fails to avert coma, even though
the pancreas contains nmnerous islands and, if coma is barely escaped,
a correspondingly high food tolerance is subsequently attainable.
These deaths, in other words, are due essentially to avoidable acci-
dents of infection, fat intoxication, etc. But with the fundamental
definition as deficiency of the internal secretion of the pancreas, a
patient can be said to die strictly of diabetes only when this secre-
tion is absolutely too scanty to support life. These are the patients
who cannot be free from active symptoms except at the price of
extreme emaciation, who cannot gain appreciably in assimilative
power, and who sometimes die finally of inanition in spite of unbroken ,
fidelity on their own part and the best skiU of the physician. Irre-
spective of any assmned functional alterations, the few autopsies
upon such patients to date have invariably shown an organic dearth
of island tissue fully siifiicient to explain the hopeless clinical severity.
The failure of treatment has thus been due to inability to replace the.
islands, which have been lost probably to some extent in an inflam-
matory process, but in most cases largely through previous errors in
dietetic management.


Certain points should also be mentioned in which the autopsy is not
decisive concerning treatment. First, absence of discoverable hy-
dropic change does not prove that the diet was suitable, since, as
mentioned above, a certain intensity of overstimulation is necessary
before this change is apt to be discoverable. Second, even extreme
dearth of islands along with absence of vacuolation does not neces-
sarily relieve the physician of responsibility for the death. It largely
absolves firom any accusation of acute death due to avoidable acci-
dents. But it has been a common mistake to feed patients slightly
beyond their true tolerance, permitting hyperglycemia and sUght
acidosis for the sake of temporary subjective comfort, and checking
more serious symptoms by occasional fasting or reduced diet. In
this way downward progress is merely delayed; the breakdown of
islands by functional overstrain occurs more slowly but just as in-
evitably. Finally, thfe most rigid treatment may be unavailing be- ,
cause the pancreatic capacity has fallen actually too low to support
life; but the cause may have been not a chronic progressive tendency
of the disease but a chronic inefl&ciency of treatment.

2. Changes Causing Diabetes. — The inflammatory origin of diabetes
carries distinct therapeutic significance. Islands once destroyed can-
not be replaced by any means now known. AU that can be hoped is
to spare the function of the remainder as effectually as possible by
diet, and thus also to provide the most favorable conditions for spon-
taneous regeneration. Acute inflammation as a cause of diabetes
offers tiree suggestions: first, the importance of sparing the weakened
function from the outset; second, the desirability of finer methods
for the diagnosis of pancreatitis, along with surgical or other pro-
cedures to mitigate the consequences; third, the chance that in at
least some cases the organic lesion may not progress further and the
subsequent welfare may be purely a question of whether the remain-
ing islands are spared or destroyed by diet. As far as diet was prop-
erly conducted, the present series supports this expectation, by
showin_g the apparent absence of inherent progressive tendency in a
considerable proportion of cases. With early eflacient dietary care, it
is to be hoped that the small number of more or less complete cures,
such as cases Nos. 40 and 76 in this series, case No. 203 of Joslin,'"

'« Joslia, E. P., Treatment of Diabetes MeUitus, 1917, p. 52.


and the extreme example of Jonas and Pepper,^' may somewhat in-
crease. Chronic inflammation makes a direct appeal for some thera-
peutic intervention to stop it. With improved dietetic control, more
attention is undoubtedly going to be paid to the diagnosis of chronic
pancreatitis and to attempts to prevent the injury which it pro-
duces as respects both diabetes and the general health. By reason
of the relative safety with which operations can now be performed
upon properly prepared diabetics, explorations for the cause of pan-
creatitis, the removal of otherwise quiescent gall stones and the
drainage of infected bile will probably become more common, especi-
ally perhaps in cases at or beyond middle Hfe. The conception of
diabetes as due chiefly to pancreatic inflammation creates a more
hopeful general view-point than before. The pathologic study thus
offers the first well grounded hope that, as constantly better control
of infectious disease is achieved, the race may some time be free of


1. The status of the islands of Langerhans as an internal secretory
organ and as the seat of the specific diabetic disturbance is now as
firmly estabUshed as any fact in physiology or pathology. In addi-
tion to the older extirpation and ligation experiments, which were
not conclusive, the new evidence consists in the production of diabetes
with large masses (up to one-third of the pancreas) of normal appear-
ing acinar tissue present when only islands are deficient, and es-
pecially in the occurrence of visible exhaustion and degeneration of the
island cells in demonstrable parallelism with variations in diet and
the course of the diabetes.

2. CUnical diabetes apparently arises regularly on a basis of pan-
creatitis, either acute or chronic; and with accurate correlation of
chnical and anatomic examinations, a geneirally logical association of-
function and structure is perceptible.

3. The explanation of the permanent lowering of assimilation re-
sulting from excess of any kind of food in diabetes is the specific de-
generation of the islands of Langerhans thus produced. Regulation
of the total diet is not merely the treatment of a symptom, but is the
essential means of preventing the principal cause of downward progress.

"Jonas, L., and Pepper, O. H. P., J. Am. Med. Assn., 1917, kviii, 1896-1897.






Gm. „,
Cal. "'





























































Online LibraryFrederick M. (Frederick Madison) AllenTotal dietary regulation in the treatment of diabetes → online text (page 67 of 76)