EFFECTS ON THE CARDIOVASCULAR SYSTEM OF FLUIDS … · EFFECT OF INTRAVENOUS FLUIDS ON CARDIOVASCULAR SYSTEM Venouspressure The venous pressure was normal before injec-tion of fluid

THE EFFECTS ON THE CARDIOVASCULARSYSTEMOF FLUIDSADMINISTERED INTRAVENOUSLYIN MAN. II. THE

DYNAMICSOF THE CIRCULATION 1

By MARKD. ALTSCHULEAND D. ROURKEGILLIGAN(From the Surgical and Medical Wards and the Medical Research Department of the Beth

Israel Hospital, atnd the Department of Medicine, Harvard Medical School, Boston)

(Received for publication February 15, 1938)

Earlier work from this laboratory ( 1 ) hasshown that appreciable increases in blood volumeoccur in man following the injection intravenouslyof isotonic or hypertonic solutions of crystalloidsin volumes of 500 to 1500 cc. It has been knownfor many years from studies in polycythemia vera(2, 3, 4, 5, 6, 7) that an abnormally large volumeof circulating blood in itself causes no significantdeviation from the normal in cardiovascular func-tion. Studies of the effects on the cardiovasculardynamics of a rapid increase in blood volume,such as occurs as a result of the administration oflarge amounts of fluids intravenously are, how-ever, fragmentary. In a few experiments Cohn-heim and Lichtheim (8), as long ago as 1877,noted a rise in both venous and arterial bloodpressures in anesthetized dogs as a result of mas-sive intravenous infusions of physiological salinesolution. Subsequently, other authors (9, 10, 11,12, 13) confirmed these findings. Meek andEyster (11), Gollwitzer-Meier (12) and Onozaki(14) reported striking increases in the cardiacoutput of anesthetized animals following the rapidinjection intravenously of large volumes of fluid.Observations in man are limited chiefly to studiesof the pulse rate, and venous and arterial bloodpressures (15, 16, 17, 18, 19). In the presentwork the effect of the intravenous administrationof crystalloid solutions on the more importantmeasurable cardiovascular functions in man havebeen studied.

MATERIAL AND METHODS

Thirty-five observations were made on 34 subjects aged17 to 71 years; twenty-six were males and eight females.Many of the patients studied were surgical cases whohad had appendectomies, herniorrhaphies, or pelvic re-pairs; in these cases intravenous fluids were administered3 to 6 hours postoperatively. Several of the subjects

1 Presented in abstract form at the May 1937 meetingof the American Society for Clinical Investigation.

were young convalescent male patients from the medicalwards; these cases were essentially normal and agreedto receive intravenous fluid for the purposes of study.One patient was studied while receiving hypertonic salinein the treatment of peripheral vascular disease; studieswere made in another patient who suffered from anginapectoris and was receiving fluids intravenously as an at-tempted therapeutic measure (20). There were no evi-dences of cardiac decompensation, hypertension, or ofmarked dehydration in any of the patients on whom ob-servations are reported. The results obtained in thesurgical patients and in the unoperated group were thesame; all studies, therefore, are treated as one group.

The minute volume output of the heart was measuredby the method of Starr and Gamble (21). Studies ofthe cardiac output before and during the intravenous in-jections were made with the patient in the postabsorptivestate and the semirecumbent position. The respiratoryminute volume, tidal air, and basal metabolic rate wereestimated in those patients on whom measurements ofthe cardiac output were made. The velocity of bloodflow was estimated from the arm to tongue circulationtime, according to the method of Winternitz, Deutsch andBrull (22). The venous pressure was measured by thedirect method of Moritz and von Tabora (23). Thepulse was counted for thirty-second periods, the respira-tions for one-minute periods. Measurements of arterialblood pressure were made by the auscultatory methodwith a mercury manometer and a standard arm cuff. Asmall calibrated spirometer was utilized to measure vitalcapacity. The effect of intravenous injections on theplasma and blood volume were studied as described in aprevious communication (1). The values for blood vol-ume changes reported here were obtained by calculationfrom the serum protein, hematocrit and estimated controlblood volumes before injection, and the protein andhematocrit values after injection (Method A of our pre-vious paper (1)). Electrocardiographic tracings weretaken with a Hindle string galvanometer. The abovemeasurements were made at frequent intervals duringand after the injection of fluids; in instances where thechanges were not great only the changes from the controllevel to the level at the end of injection are tabulated.

In analyzing the results obtained in this study, changesof less than the following magnitudes were consideredinsignificant-venous pressure, 2 cm. water; pulse rate,6 beats per minute; systolic and diastolic blood pressure,4 mm. Hg; pulse pressure, 4 mm. Hg; velocity of bloodflow, 10 per cent. By the method utilized, the volume of

401

MARKD. ALTSCHULEAND D. ROURKEGILLIGAN

fluid delivered intravenously was measurable to withinapproximately 50 cc.

In twenty-three studies a total of 500 to 1500 cc. of5 per cent solution of glucose in physiological saline was

administered; nine subjects received 1000 to 1500 cc. ofphysiological saline solution; two received 1000 to 1500cc. of 5 per cent glucose in distilled water; one received500 cc. of a 3 per cent solution of saline.

In most of the experiments venous pressures were

measured, decholin was injected for measurements of pul-monary circulation time, and blood was drawn for he-matocrit and protein measurements by means of a three-way stopcock connected with the needle inserted in thearm vein utilized for delivery of the fluid. When meas-

urements were made before the total volume of fluid hadbeen injected, the flow of fluid was stopped completelyfor two to three minutes for duplicate measurements ofvenous pressure and a single velocity measurement, afterwhich fluid injection was resumed. In some experimentsthe fluid was delivered in one arm and the antecubitalvein of the other arm was utilized for the above measure-

ments. Blood pressure readings during injection were

obtained in those experiments where the vein of only one

arm was punctured. In calculating the average rate ofinjection for a given volume of fluid, the total timeelapsed between the beginning of injection and the end ofinjection of that volume, including in many instances one

or two brief interruptions for venous pressure or velocitymeasurements or blood sampling, was utilized.

RESULTS

Pulse rate

Significant changes in pulse rate occurred in14 of 33 studies (Table I). In eleven patientsrises of from 6 to 18 beats per minute were

found; in two others decreases of 6 beats perminute were observed. The largest increases oc-

curred in subjects receiving a liter or more offluid; patients receiving hypertonic solutions ex-

hibited a greater tendency toward acceleration ofpulse than those receiving isotonic solutions.Measurements of pulse rate made one-half hourafter the end of the infusion in patients in whomrises occurred during injection revealed a tend-ency of the pulse rate to remain somewhat in-creased for this period.

Arterial blood pressure

Measurements of arterial blood pressure were

made during thirty experiments (Table I).Significant increases in systolic blood pressure

occurred in fifteen patients; in two cases the sys-tolic pressure decreased.

TABLE I

Changes in cardiovascular dynamics following the intravenousadministration of fluids

Fluid injected Increases in

Arterial Ve-Case Age blood Ve- loc- *

Pulse pressure nous ity '8iErate pres- of

:3 ~~~~~sure bloodV'Sys- Dias- flow m.tolic tolic P

cc. beatspyr per mm. mm. cm. per

min- min- Hg Hg H20 centute ute

L. C.. 39 A 500 6 -6 0.7 100. G... 64 A 550 55 6 -2 -12 200J. M . 41 A 600 23 8 6 -4 1.3 48 490tM. B.. 44 A 600 24 10 2 -6 1.2 20 320E. S5 . 48 A 600 24 4 -18 -12 2.9 560tH. F.. 20 A 600 27 -2 6 0 1.0 6 390A. F 35 A 600 40 -4 18 10 -0.1 910D. L... 44 A 600 60 6 14 8 5.4 16 570tA. S ... 43 A 700 64 -2 0 0 10.2 -4 620L. P . 32 A 1000 9 0 2 2 0.0 -3 320S. G.... 17 A 1000 11 -1.0 500A. A... 28 A 1000 15 4 2 4 -1.8 0 400W.S.. 71 A 1000 17 0 2 2 2.4 780H. S.. 52 A 1000 32 4 -2 -8 0.9 10 770G. H.. 22 A 1000 36 10 -4 -2 2.5 12 330I. F.. 27 A 1000 36 2.9 6001.O0. 17 A 1000 38 8 20 4 6.4 4 810J. Mo... 22 A 1000 40 16 8 -8 3.1 11 900T. A... 33 A 1000 43 4 3.4 880J. S... 25 A 1000 43 0 4 2 6.9 12 470L. S.... 30 A 1000 61 8 6.0 560M. A.. 28 A 1000 71 18 6 0 7.6 12 910W. S.J.. 71 A 1500 16 0 2 -4 1.2 1320L. M. §. 57 B 1000 28 0 0 6 2.4 -40S. J..... 46 B 1000 29 0 6 2 1.6 320R.S.... 58 B 1000 29 2 10 0 2.6 430tT. L.... 44 B 1000 32 4 4 -6 4.2 16 320B. G.... 42 B 1000 33 2 10 2 3.0R. C.... 37 B 1000 35 2 -2 -2 -0.1 17 60B. C.... 57 B 1000 37 6 2 O 4.1 830A. So... 52 B 1000 67 2 4 -6 0.6 13 170B. S.... 18 B 1500 58 10 4 4 0.8 0 880J. Mac.. 45 C 1000 53 -6 0 -2 6.0 990A. R.... 25 C 1500 61 -2 10 2 1.3 -17 1090G. Hy.. 25 D 500 21 -2 -2 -2 -1.2 26 480

* Changes in blood volume have been calculated fromthe estimated control blood volume, the hematocrit andplasma protein findings before injection, and the hemato-crit and protein values immediately after the injection offluids according to Method A of our previous paper (1).

t These figures represent plasma volume increases.t This study was made two days after the preceding one

in this case.§ Uremia.

A = 5 per cent glucose in 0.85 per cent saline.B = 0.85 per cent saline.C = 5 per cent glucose.D = 3 per cent saline.

Fifteen patients exhibited changes in diastolicblood pressure, six showing a rise and nine a fall.

The pulse pressure increased significantly inseventeen instances, decreased in two, and wasunchanged in the remaining eleven cases studied.Measurements of blood pressure made one-halfhour after the end of injection in those patientswhose pulse pressure showed an increase duringthe period of injection, revealed a tendency of thepulse pressure to remain increased for this period.

402

EFFECT OF INTRAVENOUSFLUIDS ON CARDIOVASCULARSYSTEM

Venous pressure

The venous pressure was normal before injec-tion of fluid in every subject. During the injec-tion of fluid at a given rate the venous pressureincreased as the amount of fluid introduced in-creased, so that the change after the introductionof 1000 cc. was on the average approximatelytwice as much as after the introduction of 500cc. (Figure 2).

Analysis of changes at the end of injection of500 to 600 cc. of fluid in thirty-two experimentsshowed greater increases in venous pressure atfaster rates of injection (Figure 1). In 17 ofthese studies the rate of injection was below 40cc. per minute; the changes in venous pressurewere not appreciable with three exceptions (Fig-ure 1). On the other hand seven of the fifteencases receiving fluids at rates of 40 cc. per minuteor greater showed increases of venous pressure

4.)

000

o 6

f)2

U.0.1z0P4)

c4

10

z

w

wL

n

U) 2w

z.

I . I

0x

0

)(* x

0

ox

x

0 20 40 60RATE OF INJECTION-CC. PER MINUTE

FIG. 1. RELATIONSHIP BETWEENINCREASE IN VENOUSPRESSURE AND RATE OF INJECTION OF FLUID INTRA-VENOUSLY; THE VOLUME ADMINISTERED WAS 500 TO

600 cc.

Dots represent cases receiving 5 per cent glucose inphysiological saline solution; crosses represent cases re-

ceiving physiological saline solution or 5 per cent glucosesolution in distilled water.

0 0

C0NiTK LEVEL

0 4

0

CC. fLUID INJECTED

FIG. 2. RELATIONSHIP BETWEENINCREASE, IN VENOUSPRESSURE, AND AMOUNTOF FLUID INJECTS

Solid lines represent cases receiving 5 per cent glucosein physiological saline solution; broken lines representcases receiving physiological saline solution or 5 per centglucose solution in distilled water.

of 2.3 to 7.7 cm. of water after 500 to 600 cc.were injected.

In several instances in which 1000 cc. of fluidwere injected the venous pressure increased tovalues above 12 cm. of water during the adminis-tration of fluid; in one subject a venous pressureof 19.2 cm. of water was obtained at the end ofinjection of 700 cc. of fluid at a rate of 64 cc.per minute.

The rise in venous pressure resulting from theintravenous administration of fluid depended toa considerable degree upon the nature of the fluidinjected. For a given volume and rate of injec-tion the venous pressure showed a tendency toincrease more after hypertonic solutions (5 percent glucose in 0.85 per cent saline) than afterisotonic solutions (Table I, Figures 1 and 2).

After cessation of injection the venous pres-sure, if increased, started immediately to returntoward the level obtaining before injection; evenin the instances where the venous pressure in-creased considerably, the values approximated the

403

l l

I I I

MARKD. ALTSCHULE AND D. ROURKEGILLIGAN

0

w0.

o4 o 1 0 2

z

z

z

MINUTES AFTER END OF INJECTION

FIG. 3. RATE OF FALL OF VENOUSPRESSURs TOWARDCONTROLLEVEL AFTER TERMINATION OF INJECTION OFFLUIDS INTRAVENOUSLY

Only those cases showing increases in venous pressureof 2 cm. or more at the end of injection are represented.

control figures in 10 to 25 minutes after the endof injection (Figure 3).

Velocity of blood flowThe arm to tongue circulation time was meas-

ured in 15 patients after 500 to 600 cc. of fluidhad been given intravenously. Significant in-creases in velocity occurred in 10 instances; in 5instances the increases were 20 per cent or more(Figure 4). The greatest increases in velocitywere observed in cases in which the fluid wasadministered at rates of injection of 20 to 45 cc.per minute (Figure 5).

A rough inverse relationship between rise invenous pressure and increase in velocity of bloodflow through the lungs was found (Figure 4).Thus, four of nine patients who showed no appre-ciable increases in venous pressure after 500 to600 cc. of fluid had been injected showed increasesin blood velocity of 20 per cent or more; on theother hand only one of six patients who did showappreciable increases in venous pressure exhibitedan increase in velocity of blood flow of thismagnitude (Figure 4).

Of fifteen patients whose velocity of blood flowwas measured more than once during the courseof injection, six showed a slower velocity of bloodflow after receiving 1000 or 1500 cc. of fluid thanafter receiving 500 or 1000 cc. (Figure 5). Inmost of the other nine patients the maximum rateof increase in velocity of blood flow was observedduring the injection of the first 500 cc. of fluid,the curve of velocity increase against amount offluid injected falling off as more fluid was in-jected (Figure 5).

Cardiac outputThe effect of the intravenous injection of 400

to 1000 cc. of fluid on the cardiac minute volumeoutput was measured in six patients (Table II).The rate of injection of fluid in these cases variedfrom 11 to 36 cc. per minute. In three patientswho showed no rise in venous pressure during theintravenous infusion the average increase in car-diac output per 100 cc. of oxygen consumptionwas 10 per cent (Table II). The average in-crease in cardiac output in the three patients inwhom rises in venous pressure occurred was ap-proximately 40 per cent. The increases in venouspressure noted in these subjects was small (TableII). The increases in cardiac output in these

600

I

40U0-Jw

z

ILl20

(0

z0

I I

Jo

00.1Q

I

I

. 1_

I

, 0

_- -4

I

x0

_ _CONrpoL LEVEL

I I I-2 0 2 4

INCREASE IN VENOUS PRESSURE-CM. H,O

FIG. 4. RELATIONSHIP BETWEENCHANGESIN VELOC-ITY OF BLOODFLOWAND IN VENOUSPRESSUREAFR IN-

JECTION OF 500 TO 600 cc. OF FLUID INTRAVENOUSLY

0

404

I

I

EFFECT IS-TkAV9NOUS7 FU-1S3- ON CARDIOVASCULARSYSTEM

CC. FLUID INJECTED

FIG. 5. RELATIONSHIP BETWEENINCREASE IN VELOC-ITY OF BLOODFLoW AND VOLUMEOF FLUID INJECTED IN

SUBJECTS IN WHOMREPEATED MEASUREMENTSWEREMADEAT VARIOUS INTERVALS DURING INJECTION

The figures in circles indicate the rate of injection.

studies resulted entirely from increases in volumeoutput per beat, since the pulse rates remainedunchanged.

The velocity of blood flow was measured inthree of the patients in whomcardiac output stud-ies were made. In one instance, the increase invelocity was proportional to the rise in cardiacoutput; in the other two cases in which the fluidswere administered at more rapid rates, the per

cent increase in velocity of blood flow was mark-edly less than the per cent increase in cardiacoutput.

The arteriovenous oxygen difference was strik-ingly diminished in the patients in whom thelargest increases in cardiac output occurred; lesserdecreases were found in the other cases.

Respiratory dynamicsThe respiratory -rate increased by more than

two respirations per minute in only two of twenty-

six studies.

TABLE II

Changes in cardiac output and related aspects of the circulationduring the injection of fluids intravenously

Cardiac output Arteriove-per 100 cc. Pulse nous oxy-

oxygen rate gen dif- C 2Wconsumed ference

.5~~~~- ?8 00

Case Cs

Be- Dur- Be- Dur- Be- Dur- : -fore ing n-fo re ing fore log .0~ 5.ini- in- in- in- in- 0,ca 3o

jec- jec- 3ec- jec- jec- 3ec- ~ - 0tio n tion craetio- tioec tio n tio n X 3 o c c

beats beats Vo7, ol - cc.itrltesper per per umes nines per cm. per

cent min- min- per per mtn- cc. cc. H20 centute ute cent cent ute

S. G. 1.86 1.86 0 68 6 5.4 5.4 11 1000 500 -1.0G. Hy.*. 1.49 1.71 15 76 76 6.7 5.9 21 400 400t -1.2 26A. A.... 1.52 1.76 16 61 58 6.6 5.7 15 1000 400 -1.8 0R. S. 1.74 2.18 25 66 68 5.7 4.6 20 800 +3.1H. F... 1.30 1.75 35 78 78 7.7 5.7 27 600 390 +1.0 6C. H.... 1.34 2.06 54 70 70 7.5 4 9 36 1000 330 +2.5 12

* The composition of the fluid administered to caseG. Hy. was 3 per cent saline; all other cases of this Tablereceived 5 per cent glucose in physiological saline solution.

t The measurement of blood volume in this case showedan increase of 480 cc. after the injection of 500 cc. of fluid;the value here presented is estimated.

The other measurements of respiratory dy-namics showed no consistent change. The res-piratory minute volume was measured in sixcases. In four cases it increased between 12 and43 per cent; it was unchanged in the other two.The tidal air was unchanged in three cases, in-creased 40 per cent in one and decreased 20 percent in two. The vital capacity, measured. infour studies, showed no change.

In no instance did the patient volunteer theinformation that he was short of breath, nor diddirect questioning in several instances reveal thepresence of dyspnea.

The oxygen consumption studied in six casesduring fluid injection was variable, remaining un-changed in two cases, decreasing 19 per cent inone, and increasing an average of 25 per cent inthree cases. Two of the patients in whom in-creases occurred became obviously restless duringthe experiment.

ElectrocardiogramStudies of the electrocardiogram by means of

continuous tracings during the injection of fluidwere made in ten patients who received 500 to1000 cc. of fluid at rates of between 23 and 71cc. per minute. In five instances no changes oc-curred, in the other five cases slight changes inthe P or T wave, or in both, were observed. The

4s05

MARKD. ALTSCHULEAND D. ROURKE<;ILLIGAN

EnU)

_ 0

_-4

o

044

w

z

0n

0

P;

z

J Jo

U

0

n

80

U,o

- z

o

U,

zU

80

U,

0

00)

off0_,

0

U

z

00

....... . < e < . ;r. . s. .... ..... u M: i w ,l: ^ ..... : : :: g v :: :; te ...... i* r %6ggA tj wr*- . t {S*W s.:ZeRe 2: e e: B 15 a. s: * £ tS a 5 B: t :^ B S t.:, A: Ss #: zBi SS .W J. h g. *. :h * h # ss :|.... 9 wW8. a as6 vAE X :- X e a itY :* W211 @er. ffi 3"2:i Bi x % *§S S._ * w f S y S <: B 5 e.: :, o . sB O* _*, B _tw5::c tt

*s g;= 3ts:{.zt y7s. #. n; ;_. . ake= b:S *.¢.eHstgr $}x. :*x:. %v9e*-., .E - eT T-a: hd # nrlr R; } # O: nc jl - ts lb S w ......

w T ZS K SSI3 EFl; Rl: !1;! 111 S #X 8 > 4 ;sX Ww, B8hE .a>2t. b: 4:iCgS x . g @ 2 . } 1: asS*:: a. e: a* fif . sEs < * .: z* r >4}* 5z 6¢ e>|S.sr 7.: *: v a sR ,., . z g _ _ * ,. < .:- * Ug: #421 #iS @SC.S ': ::. h: ss Ssr. x

:. s Sb '.313 tt b a St } e 5t 1R, *N _* trs t: K #.. tZ SS :@ tA <. ....... St 2:* :-s zs; rse s z.m xs . ..... a > - ex;Ey+|aQ1 SSa JS'iR : .' .S 1;:i e +#s ......... w.6 -f _ 8 # N ? t1 S v # ee 2i *frS UJ- ##r$ W*rs =. u sll il! 3 w t:B es .r D @*.# @*.: - .: w .x KT wEa

.> 11 2ta i$ X* 51 st wI F k :e r :: .-* St b7 fz +:q 12 g 1}8 # t St 6 U } < , $ .;* s.: .w g sst51: & a 11} s ,> . e .. .. - : se.B JsV2rt:lllr: s:R. '-: s: .i :*:<

z t; li u is .> Jl c . .: -. :: = s:s. q tI ti4i # 41 i s s - :, jR ::haj$'BIBS: _<:

4 e Jr W $11$ Z . >. .t C *. >> .z. .sX t !f t <l w D M M vS Se * t S: o:: b ow 9tS SX ' v.'S fS :: W :: b:e 1*v= saX*.>iksl . ;: ::

ffi Df * a a bS @ # 4 1P %: . .; .*: ' *.jSs 7to go=Y4 r ............ : oki: rW o aB:*, ft t} il2 tS e-#4lilRi w f ' > .i S,l Sw z q;¢* §e? & #. t S2< S.! p * Q

.e >:B 3tSSSi4#1t * w Q. aw 2 te .b B:, ,l .l>le # l t#w 2% :. 6 > r 1}gi e # #.. -. r. 3t #dlNfAlS*-SJS:& UT o: ............... 1:S b_ Jk> -tt'$< }51lS.I*d1 C iSS b t s *Er SSwI S >> Mk = 4we $,S BS XS-LS sK rlf lttr ss Sf *$ n;: os.... S. th.W 3* #Slr .> #f ............... l: -s Us 11E5Mlw*a;S$#t3flbOS :_we ............... . !-.SA!wQE e_<e ty: 2St w4wSRS 11 w* * v_ C4 f c 9i wif xr t$ B: Xsl aL sJE f s$# 4 ETS: : x:, wy $67e Ns f g * v_ es _ d Uei ISig !$S3 Sf = s,$*ws ' 8$s a *S, s<*a -4F ......... se as 1S .. os O:e>k lLs .tIS as

,. sl ss. x sw r, 1s2S X. 7 esS . d ^. tFt e Ase 3 *ze. _t:#a. wi110 S'i2 ;9#s8s:* t.:: S fb lb !11 tI #|2 q s i - 'S Will-ib# STbb:: :' :t z $tW? :1f 11 Ss nS w ¢-0 i . w _ it s1eW iRs::: - f i : v §: r 1F e ;t % t: ' < v X ES J '.1: tZqes 9 .M wS OA t q e 4 eW a :: J si e ww + + +>a# Y + S nR I #$ b R . I 11111 !J d: t t: liSt ..e 4 w * bW * S$ y v wlbV .s,eg. N a IS Ie

:S + 5 A 4 * + * P .: s: SX r } * r. a>< .t t # SS l' ......... A S:: t ts Z 4::Vwl! SF ;:.eS J r . t b +. >

'; b * 'd1 * W: 8Ia; < 8 W> + ' . ie szwCS X# F J:F * y JL * *: *; 4 1t 111 jIt t

.. st :3. tz sl- .Z: ................ ' > > >l t * BeA V :!?. e . . .e .: WG s i11 bf 2. A = b 8 f ! Su la w S. w W ._ > * 9 i 9w: :>h ers# w1S tr * #X X s,S s p

A # 21 Z # wE| 0 wXus SJ. .. > zu y wS fS SK a s: ZS : . D 1 aie-rt.> :. V 8 B r# Bw#IjFwffi IllqlS je S k-tlls:ilPl! s:*r

55 . se $wSswwe:@sY*@X r ,! N_aZ# :s5D S:SJ.3e5ydfy lfPs7 ll}5T ssZ 111 lStsill l$ *: Y:

wz wg w _ $t Tr Sl$he SS t Sgfi } \:* f Sg 1k *11 y ;i _.: lE.7ss JSwl SllJ. if B i2wS II||$ |#:E

* * rs$ <j;*S0*_ gEz f-Zle S.* SFf el eEe A _ | f 4S i rwili "_ llS 1" Jw W e:: W SF lgw siv b: UbbsAs B wWZ fe ES t s>ws: eS } s; S f:e t x Ss X. 5 ., ; s j S *.Q 2,Y W < eTs g X ¢ ^. . S B e

406

z0(f)

L-z

0LA."hiCID

EFFECT OF INTRAVENOUSFLUIDS ON CARDIOVASCULARSYSTEM

size of the P wave in one or more leads increasedin four instances; in one p2 only was increased,in two others p2 and P3, and in the fourth P1and Ps (Figure 6). Changes in the T wave inone or more leads occurred in four instances, inthree of these, changes in P wave were noted also.In one case T2 was increased, in a second T1 andT2 slightly increased, in a third T1 decreased andin the fourth T' increased in size while T2 and T3diminished. Where changes occurred, they wereusually first observed when 400 to 700 cc. of fluidhad been administered.

Four of the five patients whose electrocardio-grams showed changes received the injected fluidsat rates ranging from 30 to 71 cc. per minute.Two of the four patients whose electrocardio-graphic studies were negative received the fluidat a rate of more than 30 cc. per minute.

In two cases where electrocardiographic trac-ings showed changes during injection, tracingswere repeated two to three hours after the end ofthe injection; at this latter time the tracings wereessentially like those before injection.

DISCUSSION

It has been shown in a previous paper (1) andin Table I above that when fluids are injectedintravenously under the conditions of this studya considerable increase in blood volume occurs,and that the blood volume may not return to thecontrol level until two hours after the termina-tion of injection. The observations of the presentstudy describe the various factors concerned in theadaptation of the normal cardiovascular systemto the increases in blood volume brought aboutby fluid injections intravenously.

Whenvolumes of from 500 to 1500 cc. of fluidwere injected at slow rates (from 6 to 20 cc. perminute) there occurred appreciable increases inblood volume, as great as 1300 cc. in one case,but very little change in pulse rate, arterial bloodpressure, venous pressure, and velocity of bloodflow. On the other hand, when fluid was ad-ministered at faster rates important changes inmany of the cardiovascular measurements werefound.

When a liter or more of fluid was injected atrates greater than 20 cc. per minute the venouspressure generally became significantly increased.Several investigators have demonstrated in the

heart-lung preparation and in the anesthetized,intact animal that increase in venous pressurecauses immediate increased output of the heartdue to increased filling (9, 11, 24, 25, 26, 27).In this investigation the cardiac output was meas-ured in three instances in which the venous pres-sure was increased from 1.0 to 3.1 cm. water bythe injection of fluids intravenously. In theseexperiments the cardiac output increased from25 to 54 per cent, the increase being accomplishedthrough change in stroke volume. Presumablygreater increases in minute volume output mayhave occurred in those cases showing greater in-creases in venous pressure. The increased cardiacoutput found in this study closely accords withthe findings in animals of Meek and Eyster (11),Gollwitzer-Meier (12) and Onozaki (14). Theresults of Gollwitzer-Meier (12) and of Onozaki(14) in dogs and rabbits indicate that the cardiacoutput decreases to its control level within 15 to30 minutes after cessation of injection.

The changes in velocity of blood flow duringinjection of fluids were very variable. At moder-ate rates of injection the velocity of blood flowusually increased appreciably; with more rapidrates and larger volumes of injection, however,the increase in the velocity of blood flow wasconsiderably less than expected on the basis ofthe observed changes in cardiac output. The factthat a longer time was consumed by the blood intraversing the pulmonary circuit than that ex-pected with increased cardiac output indicates anincrease in the total cross-sectional diameter ofthe blood stream flowing through the lungs.Stewart (28) has pointed out that the slower thepulmonary circulation time with a given cardiacoutput, the greater the amount of blood in thelungs. That widely varying volumes of bloodmay be accommodated in the lungs by virtue ofthe elasticity of the pulmonary tissue has beensuggested by Blumgart and Weiss (29). Evi-dences of pulmonary engorgement sufficient tocause changes in vital capacity, or respiratorydynamics were not found in our subjects; dyspneadid not occur.

During the course of injection at moderate ormore rapid rates the pulse pressure frequentlyincreased progressively and the pulse becamebounding. A progressive and diffuse flush wasobserved in almost all the cases receiving 1000 to

407

MARKD. ALTSCHULEAND D. ROURKEGILLIGAN

1500 cc. of fluid. These findings demonstrateperipheral vasodilatation following administrationof fluids intravenously. The aforementioned ac-cumulation of fluid in the veins, as indicated byincrease in venous pressure, and in the lungs, asindicated by the measurements of velocity ofblood flow, when fluid is injected rapidly or inlarge volume must be regarded as owing to failureof the peripheral vascular bed to dilate sufficientlyrapidly to accommodate the increased blood vol-ume. That this lag in peripheral accommodationoccasions the increase in venous pressure observedunder these conditions is evidenced by the obser-vations that (1) blood volume may be increasedmore slowly to similar or greater levels withoutsignificant rises in venous pressure, and (2) whenvenous pressure does increase during rapid injec-tions it returns to normal promptly after cessa-tion of injection, although the blood volume re-mains increased. Peripheral vasodilatation is,therefore, presumably the important factor in thefinal adaptation of the cardiovascular system toincreased blood volume after intravenous in-fusions.

In general, somewhat greater changes in theblood volume and in the cardiovasular dynamicswere observed during injection of the solutioncontaining 5 per cent glucose in physiologicalsaline than when the isotonic solutions, plain physi-ological saline, or 5 per cent glucose in distilledwater, were injected. When solutions more hy-pertonic than 5 per cent glucose in physiologicalsaline are injected intravenously for therapeuticpurposes, the amounts given are usually small involume. The changes in blood volume andcardiovascular dynamics from such injectionswould, therefore, not be expected to be as greatas many of the changes observed in this study.Gibson and Evans demonstrated an increase inblood volume of approximately 200 cc. duringthe first few minutes after the intravenous in-jection of 50 cc. of 50 per cent glucose in saline(30).

The findings of elevated venous pressure andevidence of increased pulmonary blood volumein cases of this study are not interpreted as evi-dences of cardiac insufficiency. The increases incardiac output when fluids were injected rapidlywere much greater than necessary to take up theincreased fluid being delivered to the right heart;

increases in venous pressure and pulmonary bloodvolume, therefore, are not attributable to stasisresulting from myocardial insufficiency. Further,a much decreased arteriovenous oxygen differencewas found in subjects of this study with increasesin venous pressure and in blood volume followingintravenous infusions. This is in contrast to theincreased arteriovenous oxygen difference andelevated venous pressure resulting from decreasedcardiac output in patients with congestive failure.

The finding of an increase in size of the P wavein the electrocardiogram (Figure 6) is interpretedas owing to increased electrical activity of theauricles, indicating increased work (31), probablya result of increased filling. Changes in T wave,though found as frequently as those in P wave,were too variable in nature and degree to beamenable to interpretation.

The results of this study and the interpretationplaced on the findings describe the sequence ofchanges in cardiovascular dynamics after intra-venous infusions of 5.00 to 1500 cc. of crystalloidsolutions in normal subjects. As fluids are in-jected, there occurs a progressive increase inblood volume accompanied by a tendency towardincrease in intravascular pressure. Peripheralcapillary vasodilatation intervenes, and when therate of an injection is slow this mechanism ac-commodates the increased intravascular fluid vol-ume to such an extent that other changes in cardi-ovascular status are not observable. On the otherhand, during the injection of fluids at faster rates,increases in the volume of the arterial, venous,and pulmonary portions of the circulation becomemanifest. Within a short period following ter-mination of these faster injections the venouspressure decreases in spite of persistent increasein the blood volume, indicating in these instances,also, final accommodation of the increased bloodvolume through capillary dilatation. In our pre-vious communication, it was shown that the bloodvolume returns to normal within approximatelytwo hours after intravenous infusions as adminis-tered in this study (1 ). Increases in venous pres-sure result in increases in cardiac output whichpersist apparently for a period corresponding withthe duration of the increase in venous pressure.

Changes -in cardiovascular function resultingfrom rapid intravenous infusions resemble inmany ways the abnormalities in cardiovascular

408

409EFFECT OF INTRAVENOUSFLUIDS ON CARDIOVASCULARSYSTEM

dynamics owing to arteriovenous aneurysm. Inboth conditions, there is delivery of blood in in-creased volume and under increased pressure tothe heart, with a consequent rise in cardiac out-put (32). The occurrence of myocardial insuf-ficiency resulting from continued increase in thework of the heart in patients with arteriovenousaneurysm is well known.

The occurrence in elderly or cardiac patients ofpulmonary edema and of angina pectoris as a

result of the administration of fluids intravenouslyis not rare (33). It is probable that the increasedcardiac work resulting from intravenous infusionsis the principal cause of the development of thesecomplications in patients with a damaged myo-

cardium. An additional factor favoring the de-velopment of pulmonary edema is the increase inthe volume of blood in the lungs which occurs

during the intravenous administration of largevolumes of fluids. In patients with myocardialinsufficiency and peripheral vasodilatation (34)associated with increased blood volume (35) risesin venous pressure and increases in the amountof blood in the lungs would presumably occur

after smaller or slower intravenous infusions thanthose which produce these changes in normal sub-jects (11). Richards et al. (19) elaborating on

Caughey's (17) earlier work observed a decreasein vital capacity, slowing of pulmonary circula-tion time, and onset of dyspnea during rapid in-travenous infusions in some cardiac patients.These authors also noted an abnormally great andprolonged rise in venous pressure following theinjection of 1500 cc. of normal saline at a rate

of approximately 50 cc. per minute in patientswith heart disease.

Intravenous infusions which continue for a pe-

riod of several days impose conditions favoringthe development of edema, both peripheral andpulmonary, even though in such cases the fluid isgiven at very slow rates. The lowering of theplasma protein level due to plasma dilution, vaso-

dilatation due to increased blood volume, and thetendency toward increased venous pressure, by op-

erating together over a period of days, may resultin clinically perceptible edema (36).

The state of the cardiovascular system afterthe rapid injection of fluid intravenously is quitethe opposite of that obtaining in shock. Afterintravenous injections, there are observed in-

creased blood volume, increased peripheral venouspressure, tendency toward increased systolic bloodpressure, decreased arteriovenous oxygen differ-ence, flushing of the skin and bounding pulse; inshock, on the other hand, there occur decreasedblood volume, collapse of peripheral veins, reduc-tion of systolic blood pressure, increased arterio-venous oxygen difference, pallor and threadypulse.

Certain implications of the foregoing discussionare suggested in regard to therapeutics. Whenit is necessary to administer fluids intravenouslyto elderly, debilitated, or cardiac patients, the fluidinjected should be isotonic, in small volume, andinjected slowly, i.e., at rates under 15 cc. perminute. On the other hand, in the treatment ofincipient shock when blood transfusion is not im-mediately available the crystalloid solution to begiven intravenously should be hypertonic, in largevolume, and injected rapidly, i.e., at rates over30 cc. per minute.

SUMMARYAND CONCLUSIONS

1. The effects of the intravenous injection ofisotonic and of slightly hypertonic crystalloidsolutions on the venous pressure, pulse rate, ar-terial pressure, cardiac output, velocity of bloodflow, respiratory dynamics, electrocardiogram, andblood volume of normal man have been studied.

2. When 500 to 1500 cc. of physiological saline,5 per cent glucose or 5 per cent glucose in physi-ological saline solutions, were injected at rates ofless than 20 cc. per minute, very slight changeswere observed in the cardiovascular functionsstudied; the blood volume was usually consider-ably increased.

3. When these volumes of fluid were iniectedat more rapid rates considerable increases invenous pressure, cardiac output, velocity of bloodflow, and in blood volume were usually observed;increases in pulse rate, pulse pressure, and in theP wave of the electrocardiogram were observedin some instances.

4. The greater venous pressure increases oc-curred in subjects who received fluids in the largervolumes and at the more rapid rates. The venouspressure invariably returned to the control levelwithin 10 to 25 minutes after the end of fluidadministration.

5. Significant increases in cardiac output oc-

MARKD. ALTSCHULEAND D. ROURKEGILLIGAN

curred in patients in whomthe intravenous injec-tion of fluids resulted in rises in venous pressure.

6. When fluids were injected in larger volumeand at more rapid rates the increase in velocityof blood flow was considerably less than that ex-pected from changes in the cardiac output. Insome instances the increase in velocity of bloodflow was greater after the injection of 500 cc. offluid than after 1000 or 1500 cc. These findingsare interpreted as indicating an increase in pul-monary blood volume during injection. Dyspneadid not occur, and changes in respiratory dynamicswere not observed.

7. The fact that rises in venous pressure didnot persist, or even did not occur, in spite ofincreased blood volume, together with the obser-vation of increasing diffuse flush of the skin,point to a progressive peripheral vasodilatationduring the course of injection of fluids. Addi-tional evidence in this regard is the tendency to-ward increased pulse pressure observed in somesubjects.

8. The clinical implications of these findingshave been discussed.

BIBLIOGRAPHY

1. Gilligan, D. R., Altschule, M. D., and Volk, M. C.,The effects on the cardiovascular system of fluidsadministered intravenously in man. I. Studies ofthe amount and duration of changes in blood vol-ume. J. Clin. Invest., 1938, 17, 7.

2. Loewy, A., Blut und Blutkreislauf in einem Falle vonPolycythaemia rubra megalosplenica. Berl. klin.Wchnschr., 1909, 46, 1393.

3. Rover, F., Ueber Hyperglobulie. Munchen. med.Wchnschr., 1911, 58, 2791.

4. v. Bergmann, G., and Plesch, J., Ueber Hyperglobu-lie. Munchen. med. Wchnschr., 1911, 58, 1849.

5. Mohr, J., Zur Klinik der Polyzythamie. Munchen.med. Wchnschr., 1913, 60, 1739.

6. Liljestrand, G., and Stenstrom, N., Clinical studies onthe work of the heart during rest. II. The influ-ence of variations in the haemoglobin content onthe blood flow. Acta med. Scandinav., 1925, 63,130.

7. Ernst, C., Beitrag zur Frage des Kreislaufes bei derPolycythemia vera. Ztschr. f. klin. Med., 1930,114, 757.

8. Cohnheim, J., and Lichtheim, L., Ueber Hydramieund hydramisches Oedem. Virchows Arch. f.path. Anat., 1877, 69, 106.

9. Bayliss, W. M., and Starling, E. H., Observations onvenous pressures and their relationship to capillarypressures. J. Physiol., 1894, 16, 159.

10. Bainbridge, F. A., The influence of venous fillingupon the rate of the heart. J. Physiol., 1915, 50,65.

11. Meek, W. J., and Eyster, J. A. E., The effect ofplethora and variations in venous pressure on di-astolic size and output of the heart. Am. J.Physiol., 1922, 61, 186.

12. Gollwitzer-Meier, KI., Die hamodynamische Wirkungakuter Veranderungen der Blutmenge bei ver-schiedener Gefassreaktion. Arch. f. d. ges. Phys-iol., 1928, 218, 586.

13. Warthen, H. J., Massive intravenous injections. Anexperimental study. Arch. Surg., 1935, 30, 199.

14. Onozaki, N., Studien uber die Vera.nderungen derKreislaufsdynamik bei intravasalen Fliussigkeitsin-fusionen. I. Veranderungen des Minuten- undSchlagvolumens nach intravenoser Infusion vonisotonischen L6sungen. Tohoku J. Exper. Med.,1934, 24, 410.

15. Eyster, J. A. E., and Middleton, W. S., Cardiovascu-lar reactions to hemorrhage and transfusion inman. Am. J. Physiol., 1924, 68, 581.

16. Yesko, S. A., Passalacqua, L. A., and Judd, E. S.,The effect on the circulation of the injection of10 per cent glucose and 1 per cent sodium chloridfollowing operation. S. Clin. North America,1929, 9, 969.

17. Caughey, J. L., Jr., Effect of rapid infusion on ve-nous pressure: A test of cardiac reserve. Proc.Soc. Exper. Biol. and Med., 1935, 32, 973.

18. Pickering, G. W., The effect of introducing bloodfrom patients with essential hypertension -intoother human subjects. Clin. Sc., 1936, 2, 185.

19. Richards, D. W., Jr., Caughey, J. L., Cournand, A.,and Chamberlain, F. L., Intravenous saline infu-sion as a clinical test for right-heart and left-heartfailure. Tr. A. Am. Physicians, 1937, 52, 250.

20. Feinberg, S. C., The treatment of coronary arterydisease by intravenous injections of hypertonicsaline solution. Am. J. M. Sc., 1936, 191, 410.

21. Starr, I., Jr., and Gamble, C. J., An improved methodfor the determination of cardiac output in man bymeans of ethyl iodide. Am. J. Physiol., 1928, 87,450.

22. Winternitz, M., Deutsch, J., and Bruill, Z., Eineklinisch brauchbare Bestimmungsmethode derBlutumlaufszeit mittels Decholininjektion. Med.Klin., 1931, 27, 986.

23. Moritz, F., and v. Tabora, D., tYber eine Methode,beim Menschen den Druck in oberflachlichenVenen exakt zu bestimmen. Deutsches Arch. f.klin. Med., 1910, 98, 475.

24. Patterson, S. W., and Starling, E. H., On the me-

chanical factors which determine the output ofthe ventricles. J. Physiol., 1914, 48, 357.

25. Patterson, S. W., Piper, H., and Starling, E. H., Theregulation of the heart beat. J. Physiol., 1914, 48,465.

26. Straub, H., Dynamik des Saugetierherzens. DeutschesArch. f. klin. Med., 1914, 115, 531.

410

EFFECT OF INTRAVENOUSFLUIDS ON CARDIOVASCULARSYSTEM

27. Wiggers, C. J., and Katz, L. N., The contour of theventricular volume curves under different condi-tions. Am. J. Physiol., 1922, 58, 439.

28. Stewart, G. N., The pulmonary circulation time, thequantity of blood in the lungs and the output ofthe heart. Am. J. Physiol., 1921, 58, 20.

29. Blumgart, H. L., and Weiss, S., Clinical studies on

the velocity of blood flow. XI. The pulmonarycirculation time, the minute volume blood flowthrough the lungs, and the quantity of blood in thelungs. J. Clin. Invest., 1928, 6, 103.

30. Gibson, J. G., 2d, and Evans, W. A., Jr., Clinicalstudies of the blood volume. I. Clinical applica-tions of a method employing the azo dye " EvansBlue " and the spectrophotometer. J. Cin. Invest.,1937, 16, 301.

31. Pardee, H. E. B., Clinical Aspects of the Electro-cardiogram. Paul B. Hoeber, New York, 1933.

32. Laplace, L. B., Observations on the effect of an arte-riovenous fistula on the human circulation. Am. J.M. Sc., 1935, 189, 497.

33. White, P. D., Heart Disease. The Macmillan Co.,New York, 1931.

34. Crawford, J. H., Studies on human capillaries. V.Observations in cases of heart disease with regularrhythm. J. Cin. Invest., 1927, 4, 317.

35. Gibson, J. G., 2d, and Evans, W. A., Jr., Clinicalstudies of the blood volume. III. Changes in bloodvolume, venous pressure, and blood velocity rate inchronic congestive heart failure. J. Clin. Invest.,1937, 16, 851.

36. Jones, C. M., and Eaton, F. B., Postoperative nutri-tional edema. Arch. Surg., 1933, 27, 159.

411