E. A. (Edward Albert) Sharpey-Schäfer.

Text-book of physiology; (Volume v.1) online

. (page 41 of 147)
Online LibraryE. A. (Edward Albert) Sharpey-SchäferText-book of physiology; (Volume v.1) → online text (page 41 of 147)
Font size
QR-code for this ebook

dropsy or ascites may be and probably is conditioned by the increased
intracapillary pressure acting in many cases on a capillary wall already
weakened and abnormal in consequence of anaemic and diseased states
of the blood.

Comparison of the theories of Lud^wig and Heidenhain. — A
renewed examination ^ of Heidenhain's experiments, combined with a
more thorough investigation of their conditions, has persuaded me that,
so far from overthrowing the filtration hypothesis, they furnish the
strongest arguments which have yet been adduced in its favour. I may
therefore give some account of these experiments, and show how they
support Ludwig's contention with regard to the production of lymph.

Sources of the lymi^h investigated.— 1\\ dealing with the lymph flow
from the thoracic duct, it is essential to know from what parts of the
body this lymph is derived, especially since, as is well known, the
lymphatics from all parts of the body, with the exception of the
right upper extremity and right side of the neck, converge to pour
their contents into this duct. In placing a cannula in the duct,
in order to collect and measure the lymph, the ducts from the left
side of the neck and left upper extremity are ligatured. From the

^ Bayliss and Starling, Journ. Physiol., Cambridge and London, vol. xvi. p. 159 ; Star-
ling, ibid. vol. xvi. p. 224, and vol. xvii. p. 30.


hind-limbs we know that, in an animal at rest on the table, there is no
lymph flow at all. Hence the sources of the lymph are confined to the
trunk. We can, moreover, exclude the thorax and its contents, since
ligature of the thoracic duct just above the diaphragm absolutely stops
the lymph flow. Therefore, when dealing with the lymph flow from the
thoracic duct, we deal only with the lymph coming from the abdominal
viscera. As I shall show presently, the abdominal viscera, so far as their
lymph is concerned, may be divided into two groups — (1) the viscera
drained by the portal vein, and (2) the liver.

Influence of venous obstruction. — In testing the filtration hypothesis
on the lymph flow, we have to investigate whether the flow is always
proportional to the difference between the intra- and extracapillary
pressures. We may regard the extracapillary pressure as not varying
to any large extent, so that we have to see what effect is produced
on the lymph by variations in the intracapillary pressure in the
intestines and the liver. The simplest experiments on the subject
are those in which some large vessel is obstructed. Speaking generally,
we may say that obstruction of a large vein raises the pressure in
the capillaries immediately behind it, whereas obstruction of an
artery will diminish the pressure immediately in front of it. If,
for instance, we ligature the portal vein, the arterial pressure is very
little affected, while the pressure in the vein behind the ligature
rises enormously. In consequence of this, there is a large rise of
pressure in the capillaries of the intestines and spleen, so that the
spleen swells and the intestines become black from venous congestion,
haemorrhages being produced into their mucous membrane. The effect
of this ligature on the lymph flow from the thoracic duct is to increase
it four or flve times. The lymph also becomes bloody and its total
solids are diminished. The diminution in solids is due solely to a
diminution in proteids, the salts remaining the same as before ; so that
we have here an increased capillary pressure, causing an increased trans-
udation of lymph containing a diminished percentage of proteid^ — a result
which is also obtained when proteids are filtered with pressure through
dead animal membranes. The presence of red blood corpuscles in the
lymph is not a necessary consequence of a rise of pressure in the
portal vein. If a less excessive rise of pressure be produced by
ligaturing the vein, not at its entry into the liver but just below the
pancreatico-duodenal vein, thus leaving a circuitous route for the blood
to the liver through the anastomoses of this branch, an increased flow
of lymph is produced, containing less proteids than normal lymph,
but which may be quite free from red blood corpuscles.

Still more striking is the effect produced by Heidenhain's experi-
ment of obstructing the vena cava just above the diaphragm {i.e.
between the opening of the hepatic veins and the heart). The lymph
is increased from ten to twenty fold, and it is found that the lymph
obtained after the obstruction is free from red blood corpuscles and is
more concentrated than normal lymph. Thus, in one experiment of
this description, the lymph flow rose from 3 c.c. in the ten minutes
preceding the obstruction to 25 c.c. in the ten minutes after the vein was
occluded. At the same time the percentage of solids in the lymph rose
from 4'8 per cent, before, to 6 '6 per cent, after the obstruction.

What is the cause of this increased lymph flow and why is it more
concentrated ? To answer these questions we must find out first, the


source of the lymph, and secondly, the condition of the capillaiy
pressure in the organ or organs from which the lymph is derived. We
can determine the source of the lymph by a process of exclusion.
Tying the kidney vessels and lymphatics has no effect on the usual
consequences of obstructing the inferior vena cava. On the other
hand, if we hgature the lymphatics in the portal fissure which carry
off the liver lymph, we find that a subsequent obstruction has no effect
on the lymph flow, or indeed, may slightly diminish it. We must
conclude that the excess of lymph production consequent upon the
obstruction is entirely derived from the liver, and not, as Heidenhain
thought, from the intestines. The change in concentration is easily ex-
plained if we assume that, just as intestinal lymph is more concentrated
{i.e. richer in proteids) than the lymph from the hmbs, so the liver
lymph is more concentrated than intestinal lymph, or than the mixed
lymph obtained from the thoracic duct.

In order to answer the question as to the cause of this increased
production of lymph in the liver, we must investigate the changes in
the circulation jjrought about by the obstruction. On obstructing the
inferior vena cava and recording the blood pressure in the chief vessels
of the abdomen, we notice that the pressure in the aorta drops almost
at once to a third of its previous height, whereas there is a very
considerable rise of pressure both in the portal vein and inferior cava.
It is probable that the effect of the rise of portal pressure on the
intestinal capillaries is more than counterbalanced by the severe drop
in arterial pressure, so that there is a fall of pressure m the intestinal
capillaries. This conclusion is borne out by the fact that, if the
abdomen be open, the obstruction of the inferior vena cava is seen to be
at once followed by blanching of the intestines, as Heidenhain pointed
out. On the other hand, the effect of the simultaneous rise of pressures
in the portal vein and vena cava must be to increase the pressure in
the capillaries of the liver to three or four times the normal amount.
We have then, as the results of this experiment, no rise of pressure in
the portal area and no increase of lymph flow from the portal area, a
large rise of pressure in the hepatic capillaries and a very large
increase of lymph flow from the liver.

Influence of aortic obstruction. — Another experiment, on which
much stress has been laid by Heidenhain, is the one in which
the descending aorta is obstructed in the thorax. The obstruction
of this vessel is easily effected by passing an indiarubber balloon,
tied on the end of a catheter, down the right carotid artery into
the aorta just beyond the arch. The results of this obstruction on
the lymph flow are somewhat variable. In most cases the lymph is
diminished to one-half or one-third its previous amount ; in a few
cases the lymph is unaltered in quantity or even shghtly increased.
In all experiments the amount of proteids in the lymph is increased.
Now, if we investigate the state of the circulation under these con-
ditions, we find that obstruction of the thoracic aorta causes a very
considerable fall of pressure in the aorta below the obstruction and
a corresponding fall in the portal vein, whereas the pressure in the
inferior vena cava is unaltered or in some cases even slightly increased.
We must conclude, therefore, that in the intestinal capillaries the
pressure has fallen considerably below its normal limits, while in the
hepatic capillaries the pressure is very little altered or may even be


somewhat increased. Hence the only region of the body below the
point of obstruction where the capillary pressure is not much diminished
is the liver. Now we find that the liver is also the sole source of the
lymph obtained under these circumstances. If the hepatic lymphatics
be ligatured, and the thoracic aorta be then obstructed, the flow of
lymph from the thoracic duct is absolutely stopped.

These three experiments show, therefore, that the lymph x^roduction
in the organs of the abdomen is directly proportional to the capillary
pressure in these organs, and not independent of them, as was imagined
by Heidenhain.

Hydrmmia and hydrceniic plethora. — In another series of experi-
ments we find, as was predicted by Ludwig (cf. p. 288), that a marked
increase in the lymph flow is produced by a general rise of capillary pres-
sure in all the organs of the abdomen. Such a general rise of capillary
pressure may be brought about by the injection of large quantities
of normal saline fluid into the circulation, thus causing a condition of
hydraemic plethora. Under such circumstances the lymph may be in-
creased from fifty to one hundred times in amount, and may in some
cases run from the cannula in the duct in a steady stream. Now, in
hydremic plethora there are two changes in the circulation which might
possibly be responsible for the increased production of lymph — first,
the change in the composition of the blood, and secondly, the increased
pressure in the capillaries of the abdominal viscera. We can decide
which of these two factors is responsible for the increased lymph flow
by a very simple experiment. Previously to injecting 300 c.c. of normal
saline, we bleed the dog to 300 c.c, so that after the injection the total
amount of circulating fluid is the same as at the beginning of the
experiment. In this way we entirely avoid any rise of capillary
pressure, while we have diluted the blood to an even greater extent
than in the experiments in which hydremic plethora was produced.
The effect of such a simple hydrsemia is to increase the lymph flow
from 3 c.c. in ten minutes to 4 or 6 c.c. in ten minutes ; whereas, if
hydrsemic plethora were produced, the lymph would be increased from
3 c.c. to 30, 50, or 100 c.c. in ten minutes. It is evident, therefore, that
in the production of this increased lymph flow the all-important factor is
the rise of capillary pressure ; although the slight increase in the lymph
flow observed as the result of simple hydra^mia shows that, as might be
expected, a watery plasma gives rise to a transudation of lymph more
easily than does the normal more concentrated plasma.

Heidenhain s second class of lymphagogues. — In a precisely similar
manner we may explain the mode of action of the substances which
were described by Heidenhain as the second class of lymphagogues.
These include bodies such as salt, sugar, potassium iodide, etc. The
injection of a strong solution of dextrose (30 grms. in 30 c.c. water) into
the veins of an animal causes a considerable increase in., the lymph flow
from the thoracic duct. The lymph at the same time^ becomes more
watery than at the commencement of the experiment. Heidenhain
ascribes this effect to a specific excitation of the secretory activities of
the endothelial cells. The effect, however, can be explained in a much
more simple fashion. All these solutions have an osmotic pressure
which is considerably higher than that of normal blood plasma. A
solution of dextrose that should be isotonic with the blood plasma would
contain from 5 to 6 per cent, of this body. When we inject a solution




containing from 50 to 75 per cent, of dextrose, it will attract fluid from

the tissues until its percentage
is reduced to 5 or 6 per cent. ;
that is to say, 45 c.c. of fluid
containing 30 grms. of dextrose
will attract water from the tissues
until its total volume is increased
to 500 c.c. Of course this esti-
mate is merely a rough approxi-
mation at the truth, since before
the sugar has had time to attract
all this fluid, a considerable
amount of it will already have
left the vessels by diffusion. As
a matter of fact, however, we find
that injection of a strong solu-
tion of dextrose is followed in a
few minutes by a considerable
dilution of the blood, caused by
an increase in its volume. In

10' 15' 20' 25' 30' 35' 40' 4E

Fig. 40. — Diagram to show the dilution of the
blood (1 e. hydrffimic plethora) produced in
dogs (Ij^xperiments 1, 2, 3, 4) by the injec-
tion of 5 grms. dextrose per kilo, body-
weight. The ordinates represent the
volume of the blood (compared with the
normal) as indicated by percentage of
hsemoglobin. The abscissse represent inter-

vals of five minutes. The line AA, marks . . p -o i i

the time at which the injection was iinished some experiments ot von Brasoi,^

in each experiment. The dotted lines to the volume of the circulating
the left of AAi indicate the theoretical dilu- ^^^YQQ& was thus increased to twicC
tion enected by tlie volume ot nuid miected. ,, .. .,

-After J. B. Leathes. or three tmies its previous

amount; and these observations
have been fully confirmed in a series of careful experiments made by
J. B. Leathes 2 (Fig. 40).
As we should expect,
this increase in the vol-
ume of the circulating
blood is attended by a
large rise of capillary
pressure in the abdomi-
nal viscera (Fig 41), and
we have here again to
decide whether it is this
rise of capillary pressure,
or the change in the
chemical composition of
the blood, that determines
the increased lymph flow.
This question can be
solved by using the same
method that we adopted
when dealing with the
production of the in-
creased lymph flow in
hydrsemic plethora. We
can entirely olDviate the
rise of capillary pressure
if we bleed first to 300







» I







l_ - j.








- ,_




— .

6 10 20 30 40 SO 60 ininutes

In], of 40 grams dextrose

Fig. 41. — To show influence of the intravenous injection
of dextrose on the blood pressure in the abdominal
viscera, and on the lymph flow from the thoracic duct.
The upper dotted line = pressure in portal vein. The
lower dotted ]ine = pressure in inferior vena cava.
The thick continuous line = pressure in aorta. The
thin continuous line = lymph flow. The ordinates
represent venous pressure in centimetres of water,
arterial pressure in centimetres Hg, and lymph flow
in cubic centimetres per ten minutes.

C.C. and then inject a concentrated solution

1 Arch.f. Physiol., Leipzig, 1884, S. 211.

- Jovrn. Physiol., Camliridge and London, 1895, vol. xix. p. 1.



containing 18 grms. of dextrose (Fig. 42). In this case the fluid that
is dragged by the sugar from the tissues into the blood vessels only-
just suffices to make up for the previous loss of blood. No hydrcemic
plethora is produced ; there is no rise of capillary pressure, and there
is no increase in lymph flow, although an abnormally large amount
of dextrose is present in the circulation.

The fact that the immediate agent in the production of the increased
lymph flow is the hydrce-
mic plethora which suc-
ceeds the injection, explains
the point noticed by Heid-
enhain, that the efficacy
of these substances is di-
rectly proportional to their Fig. 42. — To show absence of effect of injecting dextrose
attraction for water ( Was- "^f^er a previous bleeding. Tlie description of the

• 7 .. ^ X . curves in Fig. 41 also applies to this ngure.

seranzieliungsvermogen), i.e.

to the osmotic pressure of the solution injected, and is therefore a

function of their molecular weights. A similar relation was noticed

by von Limbeck ^ to hold for the diuretic action of these bodies, which

may therefore also be possibly determined directly by the hydrsemic


The advocates of the secretory hypothesis have laid great stress on
the fact that if we analyse the lymph and the blood at different periods
after the injection of sugar, we find that the amount of this substance in
the blood steadily diminishes (even when the kidneys are cut out of the
circulation), while the sugar in the lymph gradually rises to a maximum
and then diminishes parallel with but above that in the plasma. This
was found to hold good for sugar by Heidenhain,^ for potassium iodide
by Ascher,^ and for commercial peptone by myself.''^ We are not, how-
ever, justified in concluding from these facts that the sugar, etc., have
been turned out from the blood vessels against pressure, so to speak.
As Cohnstein ^ has pointed out, the lymph flowing at any given moment
from the thoracic duct does not represent the transudation from the
blood at that moment, but is derived from the lymph that has been
formed some time previously. If we had a solution of sugar in gradually
diminishing strength flowing into a lymphatic trunk of the leg, it is evident
that this fluid would mix with the lymph in the other lymphatics, through
which it flowed on its way to the thoracic duct. Later, the solution of
sugar would have displaced practically all the lymph from these channels,
and would flow through the thoracic duct almost undiluted. It would
take, however, some considerable time to flow from the leg to the
thoracic duct, so that the outflow from the duct would represent, not the
fluid which was being injected into the leg at that moment, but the
stronger solution which had been flowing in some time previously. If
one compared, therefore, the percentage of sugar in the fluid flowing
from the duct and in the fluid flowing into the leg lymphatic at different
times after the beginning of the injection, we should obtain a curve
exactly similar to those obtained by Heidenhain after the injection of
sugar into the circulation, and regarded by him as undeniable evidence

^ Arcli. f. exper. Path. u. PharmaJcol., Leipzig, 1888, Bd. xxv. S. 69.
"^ Loc. Git. ^ Ztsclir.f. Biol., Mlincheu, 1893, Bd. xxix. S. 247.

"* Starling, Journ. Physiol., Cambridge and London, 1893, vol. xiv. p. 131,
^ Arch./, d. ges. Phijsiol., Bonn, Bd. lix, S, 350,


of secretory activity. We may conclude, therefore, that the increased
flow of lymph caused by injection of the second class of lymphagogues
is entirely due to the rise of capillary pressure thereby induced, and is in
no wise conditioned by a stimulation of the secretory activities of the
endothehal cells.

The loermecibility of the capillary wall. — The dependence of lymph
formation on capillary pressure is not the only important relationship
brought to light by these experiments. The amount and composition of
the transudation through a membrane depend not only on the pressure
at which the transudation is effected, but also on the nature of the mem-
brane. According to the permeability of the membrane, so the amount
and composition in proteids of the transuding fluid will vary. After
obstruction of the inferior vena cava, the pressure in the intestinal capil-
laries, although it probably sinks below its normal height, is yet as high
as that in the hepatic capillaries. Nevertheless, we get a very small
amount of transudation through the intestinal capillaries, and a very large
amount through the hepatic capillaries. Hence the permeability of the
liver capillaries must be very much more marked than that of the intestinal
capillaries. In the same way we may compare the permeability of the
intestinal capillaries with those of the limb capillaries. Normally from
the limb there is no flow of lymph at all, whereas a probably equal
pressm^e in the intestinal capillaries suffices to give rise to a steady flow
of lymph. If we Kgature all the veins of the leg, a lymph flow may be
set up, but such a flow is incomparably smaller than that produced on
ligature of the portal vein. We can therefore arrange the capillaries of
the body in a descending order of permeability, the hver capillaries being
the most permeable and the limb capillaries the least permeable. I have
already mentioned how, on flltering solutions of proteid through various
membranes, the percentage of proteids in the filtrate increases with the
permeabihty of the membrane. As we have seen, exactly the same
thing holds good for the capillaries in the body. The lymph in the
limbs, the filtrate through the impermeable limb capillaries, contains
only from 2 to 3 per cent, proteids ; that from the intestines contains
from 4 to 6 per cent, proteids ; while that from the permeable capillaries
of the liver contains from 6 to 8 per cent, proteids — ^in fact, almost as
much as the blood plasma itself. It is conceivable that we might alter
the amount of lymph in any organ by changing, not the intracapillary pres-
sure, but the filtering membrane, i.e. the endothelial wall of the capillaries.
Such a change can be brought about in the body by various means.
Thus the permeability of the limb capillaries is considerably increased as
the effect of any local injury, such as that caused by plunging the limbs
into water at 56 C. for a few minutes. Cohnheim^ pointed out that if a
cannula be placed in one of the lymphatics of the foot, and the foot be
then scalded in this manner, in a few minutes the lymph begins to flow
spontaneously from the cannula. The lymph which is thus produced is
much richer in proteids than is lymph from a normal limb. Moreover,
as Jankowski - showed, the amount of lymph flowing from the foot can
now be varied within wide limits by altering the pressure in the capil-
laries, either by ligature of the vein or artery, injection of salt solution,
or production of vasomotor paralysis. By this scalding, in fact, we may
reduce the limb capillaries to the condition of hver capillaries.

^ Vircho7j)'s ArcMv, 1877, Bd. Ixix. S. 516.
^ Ibid., 1883, Bd. xciii. S. 259.



Heidenhaiii s first class of lym2Jhagogues. — We are now in a position to
discuss the mode of action of the animal poisons included in the first
class of lymphagogues. On injecting a decoction of crayfish, leeches, or
mussels into the blood, the lymph flowing from the thoracic duct is
increased in amount, and becomes much more concentrated than before.
In both blood and lymph coagulability is lessened or abolished; the
blood becomes more concentrated from a loss of plasma, while the
plasma itself is less concentrated than before the injection. The blood
pressure, though generally lowered, may be unaltered if the injection be
carefully carried out ; the heart-beat is always quickened. Heidenhain
concludes that these bodies exert a specific influence on the endothelial
cells, causing them to secrete an increased amount of lymph more con-
centrated than the blood plasma.

There can be no doubt that the greater concentration of the lymph
obtained under these circumstances is due to the fact that it is chiefly
derived from the liver, since the effect of these lymphagogues on the
lymph flow may be almost abolished, if the portal lymphatics be liga-
tured previous to the injection. On investigating the changes in
capillary pressure consequent on the injection, I have found that they
are not sufficient to account for the increased lymph production. It is
true that injection of one of these bodies is invariably followed by a con-
siderable rise of pressure in the portal vein, associated with general
vascular dilatation. But this rise of pressure is comparatively transitory

Online LibraryE. A. (Edward Albert) Sharpey-SchäferText-book of physiology; (Volume v.1) → online text (page 41 of 147)