TISSUE PRESSURE(INTRACUTANEOUS, SUBCUTANEOUS,ANDINTRAMUSCULAR)AS RELATED TO VENOUSPRESSURE,

CAPILLARY FILTRATION, AND OTHERFACTORS1, 2

By HERBERTS. WELLS, JOHNB. YOUMANS,ANDDAVID G. MILLER, JR.(From the Departments of Physiology and Medicine, Vanderbilt University School of Medicine,

Nashville)

(Received for publication March 11, 1938)

Tissue pressure has often been discussed as afactor of importance for limiting the filtration offluid from the capillaries. Calculations (1, 2)have indicated that the tension in the tissues re-sulting from accumulation of fluid may attainhigh values, but available data from direct meas-urements seem to show that tissue pressure is notalways significantly elevated by prolonged venouscongestion and that even in edema the values arenot nearly as high as would be expected (3, 4, 5).The striking discrepancy between theory and ex-periment has led some authors to discount the im-portance of tissue pressure; but it should be bornein mind that the published experimental data referonly to subcutaneous and intracutaneous pres-sures, and that intramuscular pressure has notpreviously been measured in relation to problemsof capillary filtration.

1 This work has been aided by a Fluid Research Fundof the Rockefeller Foundation.

2 A preliminary report was presented before the Ameri-can Physiological Society at Memphis in April, 1937.

The present paper deals with measurements ofthe pressure in muscle as well as in the superficialtissues. The studies have been confined chiefly toobservations on normal human subjects; and ourinterest has centered mainly on tissues of the leg,because of the well known paradox of the failureof edema to develop in legs of normal individualson prolonged quiet standing. This problem hasbeen the subject of previous communications inwhich we have invoked tissue pressure as one ofthe factors limiting filtration in the leg (2, 6).

METHODS

Two methods were employed for measuring tissuepressure. In the standard method a fine needle, con-nected to a capillary glass tube, is introduced into thetissue in a horizontal position as indicated in Figure 1.Movements of minute amounts of saline into and out ofthe tissue are controlled by varying the pressure appliedfrom a water manometer. Since a change of 2 or 3 mm.H20 suffices to reverse the flow, at the point of balance,an accurate estimate of tissue pressure is readily obtained.Changes of direction of movement of the meniscus, asobserved with a cathetometer, occur within a few seconds

WATERMANOMETER

MICROSCOPEA#26 NEEDLE

All

GLASS

(HELD

CAPILLARY 0.2 MN

IN RIGID CLAMP)

FIG. 1. DIAGRAM SHOWINGMETHODOF DETERMINING TISSUE PRESSURE489

HERBERTS. WELLS, JOHN B. YOUTMANSAND DAVID G. MILLER, JR.

after the manometer pressure is altered. A correction isapplied to the manometer reading to allow for the capil-larity of the glass tube. Hundreds of measurementshave been made with this device since 1932, and we feelconfident from this experience that the method measurestrue tissue pressure, provided several obvious sources oferror are eliminated. For example, if gross bleeding isinduced the measured pressure will be abnormally highand variable and the bore of the needle is apt to becomeplugged. However, local hemorrhage is usually minimaland its results temporary. It is our practice to pull allavailable fluid into the system before withdrawing theneedle, and in successful experiments this fluid is veryslightly colored and contains only a few red cells. Inearlier experiments the precaution was taken of renderingthe limb bloodless by wrapping with a rubber bandageand removing part of the bandage from below to exposean area for measurement of the tissue pressure. Pres-sure values obtained from all tissues of the bloodless limbare the same as, or but slightly lower than, values ob-tained after the circulation has been restored. In laterexperiments we have rarely employed the rubber band-age: the needle has been inserted with the circulation in-tact. Bleeding occurs quite frequently but its signs areeasily detected so that errors are avoided simply by re-inserting the needle until a " bloodless " area is located.In the case of muscles of the forearm, however, whichseem to bleed very readily, we have attempted to meas-ure only the pressure developed during maximal volun-tary contraction of these muscles, when bleeding caninterfere but slightly with the accuracy of the measure-ment.

For the standard method, the limb must be firmly fixedso that observed movements of the meniscus in the capil-lary tube can be safely attributed only to true movementsof the fluid column. The subject lies on a rigid table,padded but without springs, and the limb to be studied isheld immobile by supporting cushions and sandbags. Theneedle assembly, which includes a reference scale notshown in the diagram, is clamped rigidly in a horizontalposition. The limb must not rest upon the table near thepoint of the measurement, for deformation of the tissuesby the weight of the limb may transmit abnormally highpressures to the region under study. Local pressure maybe changed by the needle itself, but such errors may beavoided by observation of the skin surface at the point ofinsertion. If the skin is puckered inward the local pres-sure will be elevated, and conversely if the skin is raisedby partial withdrawal of the needle the pressure will belowered even to subatmospheric levels. Lateral pressureof the needle must also be avoided. In all cases the localpressure will adjust itself to the equilibrium value ifsufficient time (an hour or so) is allowed-presumablythrough movements of tissue fluid.

Local anesthesia is rarely necessary if care is taken toseek out areas of skin and fascia which are free frompain endings. These tissues are much less sensitive inthe leg than in the arm. Sharp pain which persists andoften increases after insertion of the needle is usuallyindicative of gross hemorrhage. In a successful experi-

ment the subject may feel no pain whatever-the needlemay sometimes be inserted deep into a muscle while thesubject sleeps.

The second method allows experiments to be carriedout on the standing or moving subject, since the observa-tions do not require fixation of the limb. Several cubicmillimeters of fluid may be injected into subcutaneoustissue or muscle during the course of the experiment.The fluid flows in from the vertical manometer tube de-picted in Figure 2. It is obvious that great care must be

FIG. 2O ADHMESIVEJgW

zhi

#P22-2NEEDLE

FIG. 2. DIAGRAM oF ALTERNATIV METHODFOR USE, ON

STANDING OR MOVING SUBJECTS

employed lest the injected fluid raise the local pressurepermanently above its original level. The fluid must beable to flow freely into surrounding tissues and to returnfreely into the system, so that the equilibrium pressuremay be approached from pressures below, as well asfrom pressures above, the point of balance. Largerneedles are used than in the first method, and 2 or 3small holes are bored in each, laterally near the tip (us-ing a pointed sewing needle as drill), to facilitate thepassage of fluid into and out of the tissues. The higherthe tissue pressure the greater the possibility of errorthrough injection of an excessive amount of fluid. For-tunately the tissue fluid itself, or fluid previously injected,will flow into the system with sufficient ease to allow themarked rise of intramuscular pressure occurring duringstanding to be recorded without the necessity for repeat-ing the injections. The accuracy of the second methodhas been checked repeatedly against the standard methodin recumbent subjects. The checks have been satisfactoryover the whole range of values of tissue pressure whichwe have encountered.

The best test of the validity of the methods is perhapsthat of recording intramuscular pressure at various levels

490

TISSUE PRESSURES

of venous pressure. The fact that tissue pressure rises tothe same levels with repeated elevations to the same levelsof venous pressure, and that these values are alwayslower than venous pressure, convince us that extraneous,uncontrolled factors can play but a small part in deter-mining the measured pressures. It might be argued thatthe pressure values obtained by our methods provide anerroneous measure of true tissue tension because of thetearing apart of the tissue elements by the needle or theinjected fluid; but actually the results, in any experiment,are discarded unless it can be demonstrated that the fluidinjected locally has free access to surrounding, uninjuredtissue.

With the exception of three edematous patients onwhom measurements of subcutaneous pressure weremade, our subj ects were all normal males, ranging in agefrom 20 to 45 years. Sedentary individuals and personsof good muscular development were included. The ex-periments were conducted, in most instances, at roomtemperatures ranging from 250 to 30° C.

RESULTS

Subcutaneous pressure. Figure 3 shows thedistribution of values of subcutaneous pressureas measured in the forearm and leg of normalsubjects; in the most edematous regions of two

in 14z

j. 12zz 10-

i~

lal

0

fx 4,

2.

z)

DISTRIBUTION OF VALUES

OF SUBCUTANEOUSTISSUE

PRESSUREFOR 40

DETERMINATIONS.

NORMAL

_ CLINICAL EDEMA

PROLONGED

CONGESTION>A-> ~~~~no9 ctrAkiniliUraIMO

0 2 4 6 8 10 12 14 16SUBCUTANEOUSPRESSURECM. H&O

(CLASS INTERVALS)

FIG. 3. HISToGRAM SHOWINGDIsTIBuTioN OF VALUESOF SUBCUTANEOUSPRESSURE

ambulatory patients showing slight clinical edemaof the extremities; and in the arms and legs ofnormal subjects following prolonged venous con-

gestion produced by inflation of a blood pressure

cuff to pressures from 50 to 100 cm. H20, or byquiet standing, for periods from one to threehours in duration. Nearly half the values range

from 2 to 6 cm. H20. In general, our findingsare similar to those of Meyer and Holland (3, 4)

who report values for normal and edematous sub-jects which average 3 cm. H,0, and to those ofBurch and Sodeman (5) which range from 1 to7 cm. H,0 in normal subjects. The findings ofSmirk (7) and of Burch and Sodeman indicateclearly that in developing edema the subcutaneouspressure may rise to higher levels, especially whenthe measurement is made with the limb in a de-pendent position. They found maximum valuesof 18 cm. and 26 cm. H20 respectively. Hollandand Meyer (4) report a case of massive edema inwhich subcutaneous pressure rose to 20 cm. H20.

Wefind, in accord with Burch and Sodeman butcontrary to Meyer and Holland, that subcutane-ous pressure rises somewhat with venous conges-tion. The rise is not great-usually not morethan 1 to 3 cm. H20-but values of 11 cm. and15 cm. H2O were found in the tissues overlyingthe gastrocnemius and anterior tibial muscles re-spectively in one subject after nearly three hoursof quiet standing. Figure 4 shows that in the armsubcutaneous pressure may rise as soon as venouscongestion is initiated, and will fall at once onrelease of the congesting cuff pressure, althoughthe plethysmograph shows that the arm volumehas progressively increased nearly 100 cc. It isconcluded that a rise in venous pressure, throughits direct effect to swell the limb, may raise sub-cutaneous pressure slightly, but that the fluid fil-tered into the limb during congestion may, up tocertain limits, have little or no effect to raise thispressure further. It is probable that subcutane-ous pressure can offer no effective resistance tofiltration until edema fluid stretches the overlyingskin to near the limit of its normal elasticity.

Intracutaneous pressure. We have made fewmeasurements of normal intracutaneous pressure.The values observed lie in the range reported byMeyer and Holland (3) for normal subjects, viz.,5 to 9 cm. H2O. Following prolonged congestionof the leg, in six experiments, values from 7 to12.5 cm. H2Owere obtained, the highest figure re-ferring to pressure in the skin over the gastrocne-mius after standing 100 minutes. As in the caseof subcutaneous pressure venous congestion pro-duces an immediate rise in intracutaneous tension,rises of 2 to 3 cm. respectively having been re-corded on application to the thigh of a congestingpressure of 100 cm. H120 in experiments on twoindividuals. Wehave not measured intracutane-

491

,-t

k

HERBERTS. WELLS, JOHN B. YOUMANSAND DAVID G. MILLER, JR.

25 50 75 100 125 150 175

MINUTES

FIG. 4. INCREASE OF ARMVOLUME(CC. X 10) AND SUBCUTANEOUSPRESSUREDURING VENOUSCONGESTION

ous pressure in edematous subjects, but Hollandand Meyer (4) report that pressures in ede-matous skin are below normal levels, presumablyas the result of the breaking apart of elastic ele-ments by the accumulating fluid.

It seems evident that intracutaneous pressure isa factor of minor importance for retarding thedevelopment of edema.

Intramuscular pressure. (a) In the normal,relaxed subject in the recumbent posture. Underthese conditions intramuscular pressure valueshave been found to vary, in different individualsand in the same person on different days, from 2to 11 cm. H2O. Values below 5 cm. H2O werefound in most studies on muscles, such as the bi-ceps brachii, and the gastrocnemius, which have athin or loose fascial covering. The higher values,usually from 5 to 10 cm. H2O, refer to measure-ments on the anterior tibial and soleus, which areinvested with a tight fascial sheath.

It is probable that, as Henderson, et al. (8)suggest, variations in muscle tonus may affectintramuscular pressure of the recumbent, relaxedsubject; but our experience leads us to believe thattonus is often absent in limb muscles of normalindividuals under these conditions, and that evenunder conditions leading to postural contractionof the muscles the level of intramuscular pressureserves as a very unreliable measure of the tonus.Intramuscular pressure in the relaxed subject ap-pears to be determined by a variety of factors ofwhich the most important are the tightness of theoverlying fascia, the amount of extravascular

fluid present in the muscle, and the degree offilling of its blood vessels.

(b) Effect of venous pressure on intramuscu-lar pressure. Intramuscular pressure respondsimmediately to changes in venous pressure, evento those relatively slight fluctuations due to respi-ration. The response is much greater than that ofsubcutaneous or intracutaneous pressure in mus-cles like the soleus and the anterior tibial whichpossess tight fascial sheaths, but in the gastrocne-mius the changes are minimal. The immediateeffects of raising venous pressure by applying suc-cessive increments of pressure to a congestioncuff, with the subject in the recumbent position,or by allowing the subject to assume the erectposture, are shown in Figure 5, which presentsdata from a typical experiment. In this instancevenous pressure was recorded directly from aneedle in a branch of the saphenous vein.

It is evident that the transmission of so con-siderable a fraction of the intravascular pressureto the muscular tissue beneath the deep fascia ofthe leg provides in itself a partial compensationfor the effects of gravity on the capillary pressure.Effective filtering pressure in these muscles isless, by 20 to 30 cm. H2O, than capillary pressurebecause the vessel walls do not support the fullload of pressure. The contention of Hendersonet al. (8) that muscles give support to the veinsis therefore substantiated, but the support ischiefly due to the elasticity of fascia rather thanto the contractile pressure of the muscle alone, asHenderson has suggested; for in loosely covered

492

TISSUE PRESSURES

J.L. 12-3-36.

RECUMBENTROOMTEMP. 30C.

LEfT ANTERIOR TIBIAL MUSCLE15

5

40 60PRESSURE

80 100CM. SALINE

FIG. 5. IMMEDIATE EFFEcTs ON INTRAMUSCULAR PRESSURE OF INCREASINGVENOUSPRESSUREBY CONGESTINGCUFF OR STANDING. EFFECTS OF PROLONGEDCONGESTION (FILTRATION) AND OF GRADED POSTURAL TONE OF MUSCLES ONINTRAMUSCULARPRESSURE

muscles like the gastrocnemius and biceps brachiineither venous congestion nor the most powerfulvoluntary contraction will raise intramuscularpressure to the high levels attained in the anteriortibial by venous congestion alone. Two experi-ments on anesthetized dogs showed that intra-muscular pressure in the anterior tibial may beraised at once by venous congestion to values sim-ilar to those obtained in man. The protectivemechanism afforded by fascia does not, therefore,appear to have undergone special evolutionary de-velopment in man to favor the erect posture. Itis possible that fascia may be strengthened byexercise, as suggested by Waterfield (9), whofound that subjects in good training showed lessincrease in volume of the leg during quiet stand-ing than individuals in poor training.

(c) The relation of intramuscular pressure tocapillary filtration. Prolonged venous congestionresults in a progressive rise of intramuscular pres-sure. Final values up to 50 cm. H2O and higherhave been recorded from the anterior tibialmuscle. Figure 5 shows the time course of pres-sure rise in this muscle for thirty minutes while

venous pressure was maintained at 100 cm. H20by a congesting thigh cuff applied to the recum-bent subject. The figure also shows the elevationof intramuscular pressure at the lower venouspressures of 20 and 40 cm. H2O respectively fol-lowing the prolonged congestion at 100 cm.; andthe columns to the right indicate the levels ofintramuscular pressure in the same subject re-sulting from assumption of the erect posture, be-fore and after the period of prolonged congestionreferred to above. The effect of shifting theweight is also shown but these changes are ofinterest chiefly as they illustrate the effects ofmuscular contraction to be discussed in the sec-tion following.

Figure 6 depicts the time course of the risingpressure in the anterior tibial muscle during con-gestion produced by quiet standing.

Figure 7 presents the results of all experimentsrelating to the effect of prolonged venous conges-tion upon intramuscular pressure. It will benoted that initial and final pressures are muchlower for the gastrocnemius than for the soleus oranterior tibial.

60

o 50

U4V 40

gnU)

w 30a.

= 20-UnU)

< 10-I-z

7

0

IT

mrnm-'

I-Ir1In'-.4In'00-11

;D

IIn

00-.

0 20VENOUS

LI U_STANDING

493

HERBERT S. WELLS, JOHN B. YOUMANSAND DAVID G. MILLER, JR.

H.W. 12-7-36

ROOMTEMP. 30°C.

RIGHT ANTERIOR TIBIAL MUSCLE

STANDING. WEIGHT ON BOTH FEET. *

.

. 0

00

.11

a O<Zu-

Z~

t. -

-:rZ 0

O oXOO)V

0 10 20 30 40 50 60 70

DURATION OF STANDING MINUTES

FIG. 6. TIME COURSEOF RISING INTRAMUSCULARPRESSUREDURING

PROLONGEDSTANDING

0

24I

L

J

U)

U)

uia:

Uf)

z

DURATION

VENOUS*PRESSURE

3i i i3 § 3

cn N s e e \

o o iF F

ANTERIOR TIBIAL SOLEUS GASTROCNEMIUS

RECUMBENT STANDING STANDING (STANDING)(SITTING)

FIG. 7. EFFECTS OF PROLONGEDCONGESTIONON INTRAMUSCULARPRESSUREIN

VARIOUS MUSCLESOF HUMANSUBJECTS

60

0

z" 50U

WJ 40

U)U)Id

a. 30

-j

u 20

sz10z

0

494

.

TISSUE PRESSURES

It would appear that in the tightly covered mus-cles, as the soleus and anterior tibial, tissue pres-sure starts at a high level and increases quite rap-idly to values which are of the magnitude requiredby Starling's hypothesis to effectively oppose fur-ther filtration. In the more loosely covered mus-cle, gastrocnemius, intramuscular pressure re-mains always at lower levels, which must be rela-tively ineffective in retarding filtration. The find-ings suggest that after filtration into high-pres-sure muscular tissue has ceased, fluid will continueto accumulate in the skin and gastrocnemius mus-cle and that the volume of the leg will increasecontinuously and indefinitely during quiet stand-ing. Such a conclusion is contrary to the usualbelief, which seems to be based chiefly on evidencepresented by Waterfield (9), who found that legvolume, during standing, attained a maximumvalue within 40 minutes. Our own observationsfail to confirm those of Waterfield. Leg volumecontinues to increase as long as the subject canstand, even up to 2.5 hours. However, we haveobserved that filtration, as determined by measur-ing changes in the volume of the leg in a waterplethysmograph of the type employed by Water-field, may be temporarily masked as the result ofsharp vasoconstrictor reactions in the skin of theleg associated with the onset of acute abdominaldiscomfort, nausea, and tendency to syncopewhich invariably occur, at one time or another,during a prolonged period of standing. Intrained subjects these reactions may be post-poned to the second hour and they are less intense,but in untrained subjects they usually appearwithin the first hour and are apt to become sosevere as to terminate the experiment. The prob-able effect of these vasomotor reactions to forcea small amount of blood from the skin and thusto mask the accumulation of fluid within the tissueduring the same period requires further study,but it is our belief that such effects may have beenresponsible for the results obtained by Waterfield.Our findings are supported to some extent by theexperiments of Smirk (7) who found that thehand continues to swell for at least five hours af-ter venous congestion has been developed in thearm. Furthermore, Drury and Jones (10) wereable to develop edema in the leg of a normal sub-ject by congesting the limb for 220 minutes. Legvolume was still increasing at the end of the pe-

riod, at one-third the initial rate. These authorshave shown that a high skin temperature increasesthe rate of filtration, and in this experiment theleg was exposed to water at 360 C. Our ownexperiments have been performed at temperaturesranging from 220 to 36° C., and in all cases filtra-tion continued throughout the duration of the ob-servation. Atzler and Herbst (11) report thatleg volume continues to increase for two hoursduring quiet standing. In one experiment whichextended to three hours they report a decrease inleg volume of 4 cc., which occurred during thelast hour; and on the basis of this single expe-rience the authors conclude that leg volume willalways attain a maximum value after two hours.Such a conclusion does not seem justified by theirdata. The decrease in volume which they actuallyobserved can best be explained as due to a vaso-motor reaction.

The rates of filtration during congestion of theleg and ratios of final to initial rates, calculatedfrom our data and from figures published byDrury and Jones and by Atzler and Herbst, arerecorded in Table I.

(d) Direct and indirect effects of muscular con-traction on intramuscular pressure. As shown inFigure 8, maximal voluntary tensing of muscleswill, in some cases, raise intramuscular pressureto considerable heights. However, in relation tothe contractile force of these muscles the intra-muscular pressure is perhaps lower than onemight expect to find it. The leg muscles of ourhuman subject (H. W.) were fairly well devel-oped and it was surprising indeed that the pres-sure in the rectus femoris, for example, could notbe raised above 20 cm. H2O with the greatestpossible voluntary contraction. Figures for thegastrocnemius were not obtainable because of thepeculiar sickening pain felt in this muscle whenit contracts upon the needle, a quality of sensa-tion quite different in degree, and perhaps inquality, from that perceived in studies on othermuscles. A pressure of 31 cm. H2O has beenreported by Henderson, et al. (8) for the gas-trocnemius of a human subject during maximalvoluntary contraction.

The values of the pressure during maximal con-traction are of interest, first, in relation to thework of Anrep, Blalock, and Samaan (12) whofound that blood flow through muscles of experi-

495

HERBERTS. WELLS, JOHN B. YOUMANSAND DAVID G. MILLER, JR.

TABLE I

Filtration rates and ratios during prolonged congestion of the leg

Temperature Filtration rates | FiltrationOf ratios

Subject Conditions Dura-Water in tion -Final rateBoot Air | Intial Final Initial rate

cc. per 100 cc. per 1000 C. * C. minutes cc. per cc. perminute minute

H. W. Standing in boot; weight on opposite leg. 20-23 21 120 0.055 0.012 0.21H. W. Standing on floor; weight on opposite leg; volume,

sitting. 24-25.5 25 150 0.111 0.0135 0.12H. W.Standing in boot; weight on opposite leg. 29 30 80 0.087 0.0155 0.18H. W. Standing in boot; weight on opposite leg. Started 1

hour after preceding experiment. 36 32 150 0.064 0.009 0.14

Author

Drury and Sitting; leg in boot; congesting cuff pressure, 60 mm.Jones (10) Hg. 36 220 0.066 0.022 0.33

Atzler andHerbst (11) Standing; volume, sitting. 120 0.115 0.015 0.13

120 0.052 0.015 0.29120 0.G54 0.007 0.13120 0.055 0.018 0.24120 0.072 0.014 0.19

mental animals was much retarded and mighteven cease when powerful contractions were elic-ited either reflexly or by direct stimulation of themuscle or its motor nerve. No doubt blood flowmay be retarded during ordinary contractions inman, but it is probable that only a convulsive con-traction or cramp can completely occlude thearteries of the limb. Levels of intramuscularpressure which might conceivably be high enoughto stop the flow of blood were observed in manonly in the lower part of the soleus muscle, be-neath the thick investing layer of origin of theAchilles tendon; and in the dog, in the anteriortibial muscle during a maximal isometric contrac-tion resulting from stimulation of the sciaticnerve. Weconcur with the conclusion of Hen-derson, which was based on the single observa-tion mentioned above, that voluntary contractionsof most human muscles are not ordinarily power-ful enough to stop the flow of blood in intra-muscular arteries, for the pressures observed inmuscles are, in the main, definitely below the di-astolic level of arterial pressure.

Maximal values of intramuscular pressure areof interest from another viewpoint, for thesepressures, in most muscles, will not suffice to col-lapse even the veins, when venous pressure is

elevated during quiet standing to the usual levelsof 90 cm. H20 or higher. The veins can be onlypartially collapsed by a single contraction, andmore than one contraction will be required tocompletely empty a section of vein below a compe-tent valve. It follows also that after venous pres-sure has risen above the level of maximal intra-muscular pressure, contraction of the muscleswould be completely ineffective for pumpingblood from the veins if these vessels did notpossess their normal property of distensibility,which enables them to expand under pressure liketubes of rubber. If the walls of the veins shouldbecome indistensible, though not necessarily rigid,the venous pump would fail to act, in the erectposture, regardless of the condition of the venousvalves. Under these conditions a subject whosevalves were normal might develop some of thecirculatory changes usually associated with valvu-lar incompetence.

Our studies indicate, further, that leg volumemay or may not increase less rapidly while themuscles of the leg are held tense, as by standingwith the weight entirely on the leg in question;for, although the increase in tissue pressure re-sulting from contraction would be expected to re-tard filtration into the muscles, as claimed by

496

TISSUE PRESSURES

120

100-

0

hi

cn

U) 60.

a.

< 40,J

UU)D

< 20

z

z4

U

-I4

U

hi

tLL

w

z

Ci

hiz4-J4

tLI

ILL

wz4

u

0:-j4Uhi

c1

2 2hiz0flu

z

4

I

I~-

I-

000)

MAXIMAL VALUES OFINTRAMUSCULARPRESSURE

(VALUES FOR HUMAN

ALL FROM SAMESUBJECT)

I I I II1

90

-80

70

*60 x

50 2z

-40 'uDU)

30 wa.

20

10

- - £fI

w e : s-o e .: a.0 wX << *> - W oX zz W

x

FIG. 8. MAXIMAL VALUES OF INTRAMUSCULARPRESSUREDURING VOLUN-TARY CONTRACTIONSIN MANAND DURING MAXIMAL SCIATIC STIMULATION INTHE DOG

Waterfield (9), the prolonged contraction un-

doubtedly causes more capillaries to open up.Furthermore, it is difficult to maintain a steadytension of the muscles. Alternate contractionand relaxation, with the resulting lowering ofvenous and hence capillary pressure is apt to oc-

cur. Our experiments indicate a lower rate offiltration at first, followed by a gradually in-creasing rate; but the results were not sufficientlyregular, for reasons stated above, to warrant defi-nite conclusions.

The increase of intramuscular pressure, due tothe effect of muscular contraction, is intensifiedafter prolonged congestion as shown in Figure 5.Contraction was graded fairly accurately in thisexperiment by shifting the weight of the bodyfirst to both legs and then to the experimental leg

alone. The pressure increased in two steps as

indicated. The effect is doubtless due to the ac-

cumulation of fluid within the muscle as the re-

sult o_nration.

The elevation of intramuscular pressure result-ing from a muscular contraction may be partlyattributed to components of force exerted by thecontractile elements-the tense muscle may tendto " hug " the bone; but the fascial coveringwould seem to be of great importance for direct-ing these forces inward at points where themuscle thickens during contraction. An exactanalysis of the forces involved must await furtherinvestigation.

DISCUSSION

Special aspects of the results have been dis-cussed under appropriate headings above. It re-mains to inquire how tissue pressure can affectthe rate of accumulation of edema fluid and therate of increase of leg volume during quiet stand-ing.

It is clear that when intramuscular pressure inthe leg rises to 50 cm. H20 or higher during con-gestion at an average venous pressure of 90 cm.

L-i--JN-.E --Ir I

t

p

3.li

497

1-

HERBERTS. WELLS, JOHN B. YOUMANSAND DAVID G. MILLER, JR.

H2O or thereabouts filtration into the muscles inquestion will probably cease, for the estimatedeffective filtering pressure-40 cm. H20-is nolonger higher than the probable value for effectiveosmotic pressure. The colloid osmotic pressureof plasma proteins may be expected to rise toabout 50 cm. H2O (2) as the result of concentra-tion of the blood in filtering areas, and, allowingfor some leakage of protein through the capillarywall (13), a value of 40 cm. H20 for the effectiveosmotic pressure seems to be a reasonable figure.

Since tissue pressure in the skin, subcutaneoustissues, and gastrocnemius muscle does not riseabove 15 to 20 cm. H2O, effective filtering pres-sure (90 -20 = 70) must remain considerablyhigher than osmotic pressure, and it can thereforebe concluded that filtration must continue in theseareas as long as the subject continues to stand.If leg volume should attain a constant value withinthe first hour, as claimed by Waterfield (9) onewould be forced to conclude that fluid can be re-moved from the leg as fast as it filters from thecapillaries of these low-pressure tissues. Suchremoval might be imagined to take place bylymphatic drainage of the area, but it is very un-likely that lymph escapes from the leg at all dur-ing standing, in the absence of movements of themuscles or joints. Since we find, in fact, that theleg does continue to increase in volume for atleast 2.5 hours it would be necessary to invoke anadditional balancing factor, such as lymphaticdrainage, only if the final rate of swelling of theleg could be shown to be so very slow, relative tothe initial rate, that the retardation could not beattributed solely to the known increments ofosmotic and tissue pressures. As a matter of factthe following calculations indicate that cessationof filtration in the large mass of muscles of theleg, as tissue pressure rises in these high-pressureregions, is chiefly accountable for the low residualrate of swelling of the leg.

The leg of a cadaver, a well developed malesubject, was dissected and the parts weighed.The skin weighed 0.33 kgm. (18 per cent of thetotal weight of skin and muscles), the gastroc-nemius weighed 0.36 kgm. (19 per cent of thetotal) and all other leg muscles weighed together1.185 kgm. (63 per cent of the total). The sub-cutaneous tissue was neglected, for the number ofopen capillaries available for filtration is probably

minimal in this tissue. The tissues of the footwere also neglected, on the assumption that therelative weights are similar to those of the leg.Assuming that, aside from differences in tissuepressure, filtration occurs at the same rate, perunit weight of tissue, in skin and all muscles; thataverage capillary pressure throughout the leg re-mains at 90 cm. H20 during standing; that effec-tive osmotic pressure rises from an initial value of30 cm. to a final value of 40 cm. H20; and thatinitial and final values of tissue pressure are: inskin, 5 and 15 cm. H2O; in the gastrocnemius, 5and 20 cm. H2O; and in all other muscles 25 and50 cm. H20 respectively; one can calculate, usingthe relative weights of these tissues as deter-mined, that the ratio of final to initial filtrationrates will be 0.27. If tissue pressure did not riseabove its initial value in the high-pressure musclesthe ratio would be 0.65, showing that the rise inintramuscular pressure which is actually observedhas a marked effect to retard filtration. Sincevariations in skin temperature are known to affectprofoundly the rate of increase- of leg volume, wemay profitably inquire to what degree a relativechange in rate of filtration in skin alone will affectthe calculated ratios of final to initial filtration inthe leg. If filtration occurs at twice the rate inskin as in muscle, the ratio rises to 0.33; if filtra-tion in skin starts at one-half the rate for anequivalent weight of muscle, the ratio drops to0.23; if the cutaneous rate is one-fourth the stand-ard rate for muscle, the ratio is reduced to 0.21;and if we can assume that filtration starts in theskin at the standard rate but stops entirely duringa phase of vasoconstriction affecting this tissuealone, the ratio falls to 0.14. It is of interest tonote that the range of calculated ratios (0.14 to0.33) very nearly covers the range of observedratios (0.12 to 0.33.) presented in Table I.

It is not our intention to place undue emphasison the particular values of the calculated ratios.The calculations are merely helpful to indicatethe approximate relationships of the knownvariables which affect the rate of swelling of theleg during quiet standing and to show that, in allprobability, the observed changes in osmotic pres-sure and tissue pressure are sufficient to accountfor the changes in filtration rates, if a reasonableallowance is also made for changes in the filteringarea incident to vasomotor reactions.

'498

TISSUE PRESSURES

SUMMARY

1. Subcutaneous and intracutaneous pressures

are never greatly elevated in normal subjects fol-lowing prolonged venous congestion of the legand in most cases the rise is insignificant.

2. The pressure in those leg muscles which are

tightly covered with fascia rises to 50 cm. H2Oor higher during prolonged venous congestion.Pressures of this magnitude probably suffice tostop filtration into these muscles during quietstanding. Pressure in the gastrocnemius, whichis loosely covered, does not rise above 20 cm. H2Oduring congestion.

3. Intramuscular pressure is affected independ-ently by several factors, of which the most impor-tant are: the tightness of the fascia, the amountof intravascular and extravascular fluid present,and the components of contractile force of themuscle. Intramuscular pressure cannot be usedas an index of muscle tonus unless all of thesefactors are controlled.

4. Maximal voluntary contractions of variousmuscles elevate intramuscular pressure to valuesranging from 10 to 118 cm. H2O. As thesevalues, in most muscles, are lower than the levelof diastolic arterial pressure, blood flow throughthese muscles cannot be stopped during ordinarycontractions. Relations of intramuscular pres-

sure to the venous pump mechanism are discussed.5. Leg volume continues to increase indefinitely

(at least for 2.5 hours) during quiet standing.In each case the ratio of the final rate of filtrationto the initial rate is of the magnitude to be ex-

pected on the assumption that filtration ceases inhigh-pressure muscles but continues indefinitelyin the skin and gastrocnemius, the low pressure

filtering areas. It is not necessary to assume thatfluid leaves the leg through lymphatics duringquiet standing.

BIBLIOGRAPHY

1. Landis, E. M., and Gibbon, J. H., Jr., The effect oftemperature and of tissue pressure on the move-ment of fluid through the human capillary wall.J. Clin. Invest., 1933, 12, 105.

2. Youmans, J. B., Wells, H. S., Donley, D., and Miller,D. G., The effect of posture (standing) on theserum protein concentration and colloid osmoticpressure of blood from the foot in relation to theformation of edema. J. Clin. Invest., 1934, 13,447.

3. Meyer, F., and Holland, G., Die Messung des Druckesin Geweben. Arch. f. exper. Path. u. Pharmakol.,1932, 168, 580.

4. Holland, G., and Meyer, F., Der Gewebsdruck beimOdem. Arch. f. exper. Path. u. Pharmakol., 1932,168, 603.

5. Burch, G. E., and Sodeman, W. A., The estimation ofthe subcutaneous tissue pressure by a directmethod. J. Gin. Invest., 1937, 16, 845.

6. Youmans, J. B., Akeroyd, J. H., and Frank, Helen,Changes in the blood and circulation with changesin posture. The effect of exercise and vasodilata-tion. J. Clin. Invest., 1935, 14, 739.

7. Smirk, F. H., Observations on the causes of oedemain congestive heart failure. Clin. Sc., 1935-36, 2,317.

8. Henderson, Y., Oughterson, A. W., Greenberg, L. A.,and Searle, C. P., Muscle tonus, intramuscularpressure and the venopressor mechanism. Am. J.Physiol., 1935-36, 114, 261.

9. Waterfield, R. L., The effect of posture on the vol-ume of the leg. J. Physiol., 1931, 72, 121.

10. Drury, A. N., and Jones, N. W., Observations uponthe rate at which oedema forms when the veins ofthe human limb are congested. Heart, 1927, 14,55.

11. Atzler, E., and Herbst, R., Die Schwankungen desFussvolumens und deren Beeinflussung. Ztschr. f.d. ges. exper. Med., 1923, 38, 137.

12. Anrep, G. V., Blalock, A., and Samaan, A., The ef-fect of muscular contraction upon the blood flowin the skeletal muscle. Proc. Roy. Soc. London,s.B., 1933-34, 114, 223.

13. Landis, E. M., Jonas, L., Angevine, M., and Erb, W.,The passage of fluid and protein through the hu-man capillary wall during venous congestion. J.Clin. Invest., 1932, 11, 717.

499