THE CARDIOHEMODYNAMICEFFECTSOF VENOUSCONGES-TION OF THE LEGS OROF PHLEBOTOMYIN PATIENTSWITH ANDWITHOUTCONGESTIVEHEARTFAILURE1

BY WALTERE. JUDSON, WILLIAM HOLLANDERJ2J. D. HATCHER,' MEYERH.HALPERIN, ANDIRWIN H. FRIEDMAN4

(From the Robert Dawson Evans Memorial Department of Clinical Research and PreventiveMedicine, Massachusetts Memorial Hospitals, and the Department of Medicine, Boston

University School of Medicine, Boston, Mass.)

(Submitted for publication May 18, 1954; accepted by December 29, 1954)

The purpose of this study was to clarify thecardiohemodynamic responses to acute reductionin the "effective circulating blood volume" in man.McMichael and Sharpey-Schafer found that thecardiac output in normal subjects fell during ve-nous congestion of the extremities (1). Theyattributed this reduction in cardiac output to adecrease in the right atrial pressure. Their ob-servation that right atrial pressure usually fallsduring such acute reductions in the "effectivecirculating blood volume" was confirmed by War-ren, Brannon, Stead, and Merrill (2), but theselatter workers did not usually find an associateddecrease in cardiac output. They therefore sug-gested that the cardiac output response is inde-pendent of changes in right atrial pressure, atleast within physiological limits. However, ob-servations by Howarth, McMichael, and Sharpey-Schafer (3) on the effects of phlebotomy and ofvenous congestion of the extremities in patientswith congestive heart failure supported the con-cept that the level of right atrial pressure is animportant factor in determining cardiac output.The studies reported herewith differ in certainaspects from all those just mentioned.

MATERIAL AND METHODS

The patients in Group I had neither clinical nor hemo-dynamic evidence of cardiac insufficiency at rest or af-

1 This investigation was supported in part by a grantfrom the National Heart Institute of the National In-stitutes of Health, U. S. P. H. S.

2 United States Public Health Service Fellow in Medi-cine, Evans Memorial, Massachusetts Memorial Hospitals.

8 Formerly Fellow in Medicine, Evans Memorial, Mas-sachusetts Memorial Hospitals. Present Address: Queen'sUniversity, Kingston, Ontario, Canada.

4 Formerly United States Public Health Service Fel-low in Medicine, Evans Memorial, Massachusetts Mem-orial Hospitals. Present Address: East Meadow, NewYork.

ter exercise. They included two normal subjects, onewith essential hypertension, and two with hypertensivecardiovascular disease. Group II-A contained patientswith non-valvular types of heart disease with congestivefailure and included three patients with hypertensivecardiovascular disease and two with arterioscleroticheart disease. Group II-B contained patients with val-vular types of heart disease and congestive failure andincluded five with rheumatic heart disease and one withsyphilitic heart disease. All of the patients in GroupII-A and B were considered to be in congestive failureon the basis of their clinical cardiovascular symptomsand signs and their abnormal hemodynamic responsesduring exercise. Furthermore, all of the patients weretaking digitalis, and several were receiving mercurial orother diuretics for treatment of their congestive failure.Group III consisted of patients with pulmonary emphy-sema, two of whom had cor pulmonale with congestiveheart failure and two of whom had pulmonary hyperten-sion with no evidence of right ventricular hypertrophy orfailure.

PROCEDURE

The patients were studied in the post-absorptive statewithout any preliminary sedative medication. A double-lumen intracardiac catheter was inserted into the rightheart so that simultaneous records of the pulmonary ar-terial and right ventricular pressures could be obtained.An indwelling needle was maintained in the brachialartery. All pressures were measured with electromanom-eters5 and recorded by direct-writing oscillograph; thezero point of reference for all pressures was 10 cm. abovethe back of the supine subject. Mean pressures were de-termined by electrical integration. Using criteria whichare comparable with those in Cournand's laboratory, re-ductions in either the right ventricular diastolic or meanpulmonary arterial pressures of 5 mm. Hg or more wereconsidered physiologically significant, while changes inthe mean systemic arterial pressure of 15 mm. Hg werealso judged physiologically significant (4).

The resistances were calculated, according to Gorlinand his co-workers (5), by the following formulae:

The "total pulmonary" resistance (actually, total

6 Sanborn Co., Cambridge, Massachusetts.

614

CARDIAC RESPONSESTO VENOUSCONGESTIONAND PHLEBOTOMY

0N

c0N N"

00- 0 0U) 00410'I1°100 180~I" _ IN

N N N N N

iiool .II oo SI 0 I I I10 N1-1 001

I I 1--1-1 .1-1--N

I-I I II I -I _) I)co 0 00

_I_I I__I__I_

U4 -f@0 41 o %4 ) U)0% C0%0% 0%a %%0% 0 %0 %%0

O.oW 0 oW okov oo ok

0. loc 0 101 al0 if) 0 o10N~aN0 O N 0%N U) 0 %q0o0

N tN N N" )14) if) NN°.1^in if)-

SI %% 4Nn N N )0

o S I I__I o I '" oI0 0 01 1 .t- 1^^) %O 1O co^ Ole1f>

0 I O OI . . , ,I

4i - -3 va|2N3 I..00 000 0000 1

Z) '1t 1. 00

.44

r-

P.8.

NC4

0@0

.00%.

0Go@0

00Nq -40%0

0co

00I

U)

if (if; %1 UolW ( Wolo 18o ooCR 9 C1 8i Iq a , o|¢|¢stiq C

4'_ 0 0@oIo0 0NI N NN@001@0INI_

~~0%@00 0% 0-

1w^o001o0 U1).") U00(4C4(4 .4~~@ -- ".

00 0 0 0 1) 10 0--1--

0 00 0 0 N 0 0 @00 0 UU)wU u)

U)U)U)N N '.' 'N~co - @040 @040 @0

)U 0 U) U) co@@0N 0000-N-N- NNN N

I N co0 o@0 0% NN U) U) in W) in

00ol 0 0o ) i n N 0col c co co coN Nou) N N W)N_O 'IO 0'0 10 0

VUlq44 if) 04) if ) N% 0U

1"1""1" 1" 1" 1 ~"1- I- 1- 1-I

@0eeoull@0@0 0% 0% 0%10

% 0%co

UIN

0%_ NI ) 4' N U) 4' @0_ _I s I

No f 0 OVe4oo N* 4101 1@01if) if1 ) 1t

-(4loo lo o NOV_loo 0U4'4 4'4' 4'

W%@0 0 0% N 0-' U0% 0 t

@0N0%u 4' N No vm v IN 44'o 10o 0

U)U N N 4' 4' 1°NN U)fi1

NN@0 ) e%. )4'U 0)

0 0I I 0 0 0 1 0 V- 0L 0

I-_

aIVwIN00I0 *eIe Ib 0t10

I T.r L~~~~~~~~~~~~~~~.z

.ell

_t

I.xI e

.0

@0cAL100

I4'

615

.4

W4 -I

013

CI)...,

0

II'

I.-o, 0

I .0~I 'A

0'0GD3

~ 3

.0...

1'2°|,44

1-GD

'X0

I ..

I c 0GD4

CI ) * :4I*4-t

-:434 Go*0

° 18

W. JUDSON, W. HOLLANDER, J. HATCHER, M. HALPERIN, AND I. FRIEDMAN

0o

m-4

0' 0 oI 0101eq) eq III

00

t1 0o I In 0 I0 0 co t- 0

1+lelools~~~~~~~~~~~~~i1q1 1 1 lolo e 10 1-°°1 0-00- eq0 10%0 '0'0 1.+0*-C( U )( N N co

I telualEglo lole I co o colco lu elo&O '000In '0 a 0 0 (90400

1>>1b1--1° 1v°1" 1^°1^° 1 -1°'IO co

000 % ) i if)U U) 0' 00 00Ul) U) U) 0 0 0U) if) UU)

0 0 0 0 - -0- %I WI I I -I ' 'I I

II I l+FI 1-1

UUUO101U) U U I) 00 0 1

*2121qe¢1< (1- (9(9 1- 1-1

c(00 0 100%0% 0 0 U) U) 0a

N,N

U) n0

Une) eq

'0 '0 U)U)a1eq cola141 1%0 >1 1> 10000 U)1b010U)>10

11 I n t-

inv101 c+1Ao loiv 11couaa

0 000006dr 6r 6 44 u00(9 0 Ii "0"r:

I__ a InoI1 IIco %O 0 " U) 000000

I N1 1Se1r1t- 14 1e1 1eo 0 ocoU)0 00 )0, e qo%'00 0 0t UI, U)I

I I n 1 .

0 00% 00%%0 0%0 0%%0000%00%% 00~~~~~~.

0

to be|4|SX erx|Qvs> ->0 M z 0 ¢ !000 0 000 0 ..! 0

00000.00-0 ~~~' ~' U) ~ (900 000'-~ 0

'0

Uf)a 1

0

co co° lN

000 oN0'0 000co o000

0 000 U) U)0 0 00Y)t>°g | o o ot t-t0-8

-- -1 1-1- 1-1-

I) o o U) 0) WI)eq eq Cq C'4 e4

eq q U) eL0 W"_

000 S 0 O0 0 co oO

1 'U) I-q - 01

U4 )

U)U )0 eq 4 U) U) 0)U) '0 - 04e '0 10.

o N U) U) U)

eq U U) 0 4 U) v

U) In t- eq U)U U) 04 co

UoU)o04 0- -) U) 00000 C'4U) 0 - U)0o.- '0. '0x U) '0s U)

1- l

0

0 0 F. 0° 0.10 0 0 .0~0 0 0,UU U Uo l

0 0 0

e ~ E0U2- 2Sv 04UoZ6 uUZ

.:4

=: 0

W)

616

0LI)NO

0

0%

0*

L

m

i

IX

*s:c

rw. oW C4lo l

m me l1* 10 0

k)

3I43.h3e:I

tNa.kt

144

"I9

5

4i 'o,q

O.;t- -4

1 2:4.-, .Q

C4

W)

CARDIAC RESPONSESTO VENOUSCONGESTIONAND PHLEBOTOMY

resistance opposing the right ventricle) was obtainedfrom the following equation:

R = co X 1,332 dynes seconds cm.7

The "total systemic peripheral" resistance (actually,total resistance opposing the left ventricle) was obtainedsimilarly:

R = co X 1,332 dynes seconds cm.-

where PA. = pulmonary arterial mean pressure, mm. HgBA. = brachial arterial mean pressure, mm. HgCO= cardiac output, ml. per second

1332 = conversion factor from mm. Hg to dynesper cm.2

After preliminary practice, one or two control cardiacoutputs were determined by the direct Fick method.Expired air was collected for two minutes in a Douglasbag while simultaneous blood samples were obtained at aconstant rate from the brachial and pulmonary arteries.Immediately before and after the Fick procedure, pres-sures in the pulmonary artery, right ventricle, andbrachial artery were determined. The volume of in-spired air was measured by a Tissot spirometer, whichcontained a supply of outdoor air. Oxygen and carbondioxide concentrations in the expired air were determinedby the Haldane method (6) with duplicated determina-tions required to agree within 0.03 per cent. Bloodsamples were analyzed for oxygen content, capacity, andsaturation by the method of Van Slyke and Neill (7)with duplicate determinations required to agree within0.10 vol. per cent. Possible slight errors in blood oxygencontents due to dilution in obtaining blood samplesthrough the catheter were corrected as follows:

Corrected venous oxygen content

= Observed venous oxygen content X art. hct.ven. hct.

In fourteen patients the cardiohemodynamic measure-ments were repeated 7 to 68 minutes after application ofvenous congesting cuffs high on the thighs at 70 mm. Hg.In six patients similar measurements were made from 3to 20 minutes after the completion of a 450 to 750 mil-liliter phlebotomy. The venesections were accomplishedover a period of 10 to 40 minutes. A few patients inwhom the oxygen consumption during the control andprocedure periods did not check within 10 per centwere excluded from the study. Likewise, patients inwhom the cardiac output during the control periods va-ried more than 10 per cent were also excluded from thestudy (4).

RESULTS

The cardiovascular responses to venous con-gestion of the extremities and phlebotomy aresummarized in Table I and the means are shownin Table II with a statistical analysis.

E

i4i0Fe E t

Au

0

000%%MN

-H-H-H0004

coo

000t-.IC'0 0C-U U) C'4

-H -H-H000-0%'.1-

0.0

0- It001-H--H

C.0o Cs- v

+

+ :

I+0 0 A 04

3 > E | " ° |w. oa o |+AN N

0 N00 0 00o 0 0 A

~~~ ~ I +

s i~H C -H -H-H -

Am

14 ChaNs acN_n C

a n°E_ N o|+o

.CO j,0 -HbH -H -H-HeCNNe~~~~ t.VCU a

ci2tte.~-G 0oo

I __iw I

-IV

Joo 4i-il40 41-H J'0Hc

0 °9 | Ot V) V | VM0 N O | 0 O

. . 0MN '0n

+I +

F. 'p i coo-' Go

> . H-H41 -H-H-H41 -H C!

C.'0 C-O~~~~+

bo -H -H-H W, FH-H4i_____ C4Cm I CC'4

*

W t' '0co C;.C

Ch W) W) aU004 co

I+

-., 00

00

4Ut'

4,

0 0

0-

eo 0!U

1-04)uv ciu

Cl cis EV UZ.. = 4)to

r.-tN .

.- be'aoui

cl'..4

I -

-0 cid0006

a g

F6

617

0

la

laCUco

mvC.IV

CU

*k

l

L. L-

W. JUDSON, W. HOLLANDER, J. HATCHER, M. HALPERIN, AND I. FRIEDMAN

Group I-Patients without heart failure

A. Cardiac output and oxygen measurements.All five patients showed falls in cardiac output,four of which were greater than 10 per cent.The mean decrease in cardiac output was 1.53liters per minute which was highly significantstatistically. This was associated with a mean

increase of 0.66 vol. per cent in the arterio-venousoxygen difference, a change which was also highlysignificant. The oxygen consumption, arterialoxygen saturation, hematocrit, and pulse rate didnot change significantly.

B. Vascular pressures. In two patients (M.R., I. R.) in whom it was measured, the end-dia-stolic right ventricular pressure did not changemeasurably from the control normal values duringcongestion of the limbs despite a decrease in thecardiac output. Only one patient (W. D.) showeda physiologically significant decrease in the pul-monary arterial pressure. The group as a wholeshowed no significant change in the pulmonaryarterial pressure. There was a tendency for theperipheral arterial pressure to decrease, but thechange was not statistically significant for thegroup. Indeed, only one severely hypertensivepatient, W. D., had an appreciable drop in bloodpressure.

C. Vascular resistance. Both the "total pul-monary" and "total systemic" resistances showedslight but not significant increases. The directionof these slight changes was such as to tend tomaintain constancy of pressure with the changein cardiac output. In only one patient, W. D.,who had a considerable fall in arterial pressure

did a correspondingly large change in the systemicvascular resistance occur.

Group II-Patients with heart failure

The decompensated patients with valvular andnon-valvular heart disease responded similarly tostimuli which reduced the "effective blood vol-ume." The averages presented in Table II includeboth types of patients.

A. Cardiac output and oxygen measurements.On the average there was a slight but statisticallysignificant increase in cardiac output of 0.28 litersper minute. This change was associated with a

mean decrease of 0.81 vol. per cent of the arterio-venous oxygen difference, which was also signifi-

cant statistically. The direction of these responseswas opposite to that of the compensated patients,and the difference between the responses of thetwo groups was highly significant statistically.The oxygen consumption, arterial oxygen satura-tion, hematocrit, and pulse rate did not changesignificantly.

B. Vascular pressures. Initially the right ven-tricular end-diastolic pressures were elevated inalmost all the patients. Venous congestion of theextremities or venesection produced a mean de-crease of 6 mm. Hg in the right ventricular dias-tolic pressure in the group, a highly significantchange statistically and physiologically. The pul-monary arterial systolic, diastolic, and mean pres-sures likewise showed significant decreases of asimilar order of magnitude. The pulmonary ar-terial pressures in these patients changed in thesame direction as in the compensated group butto a significantly greater degree. On the averagethere was an 8 mm. Hg decrease in peripheralmean arterial pressure, a change which was notstatistically significant.

An interesting illustration is M. W., a patientwith mitral valvular heart disease and functionaltricuspid insufficiency. This patient had the larg-est increase in cardiac output of the entire groupafter venesection, and this was associated with thegreatest reduction in mean right atrial pressure.After phlebotomy the mean pressure and the am-plitude of pulsations in the right atrial pressuretracing, as observed in Figure 1, were reduced,possibly due to a decrease in the volume of bloodregurgitated through the tricuspid valve. Thecardiac output had increased 53 per cent.

However, not all the patients in the group whohad increases in the cardiac output during thesemaneuvers had decreases in the vascular pres-sures. Conversely, not all the patients who hadsignificant reductions in the right ventricular dias-tolic and the pulmonary arterial pressures hadsignificant increases in cardiac output. Further-more, in those patients in whom there were si-multaneously measured increases in cardiac outputthere was no consistent correlation with the de-gree of reduction in the right ventricular diastolicand pulmonary arterial pressures.

C. Vascular resistances. The "total pulhno-nary" resistance decreased, on the average, 14 percent, a highly significant statistical change for the

618

CARDIAC RESPONSESTO VENOUSCONGESTIONAND PHLEBOTOMY

*FORE PHLOOTOMY AFTER PHLEDITOY

r.-.! I. LLJ.. .1: 1 .; 1 1 L IT;

goo-1-1

sE I tio-.,0e .M

40-0

5UM 0

_ _ _ i_ I_ _ Lp_iu_

aI I

2

_ GO

M.W *, 5,W.

FIG. 1. RECORDINGOF THE RIGHT ATRIAL PRESSUREIN PATIENT (M. W.)WITH RHEUMATIC HEART DISEASE AND FUNCTIONAL TRicusPI INSUFFi-CIENCY SHOWINGTHE REDUCTION IN MEANPRESSUREAND AMPLITUDE OF

PULSATIONS AFTER PHLEBOTOMYThis change may be due to reduction in degree of regurgitation. This

patient likewise had a marked increase in cardiac output.

group. The total systemic resistance also de-creased 16 per cent, but the change lacked statisti-cal significance. The vascular resistances changedin a direction sich that the blood pressure tendedto remain at a constant level with the simultaneousincrease in cardiac output. The decreases in re-

sistances in this group contrasted with the in-creases in the compensated patients. The differ-ence between the responses of the "total pulmo-nary" resistance of the two groups was highlysignificant statistically.

Group III-Patients wuith pulmonary emphysemawith or uithout heart failure

The responses to venesection of two patientswith heart failure due to pulmonary heart diseasewere different from those with heart failure dueto forms of heart disease generally classified as

"low output failure." Therefore, these patients as

well as two others with actual or potential cor

pulnonale without heart failure are presented sep-

arately from the other groups. The small numberof patients studied rendered statistical analysis ofthe results impractical.

The two patients (D. K., C. Di.) in this group

without heart failure had elevated pulmonary ar-

terial pressure, distinguishing them from the pa-

tients in Group I. Both of these patients had re-

ductions in cardiac output during limb congestion(D. K.) or after phlebotomy (C. Di.). However,in patient C. Di., the decrease in cardiac outputwas proportionate to a fall in the oxygen consump-

tion without an appreciable change in the mixedarterio-venous oxygen difference. In this patientthere was a fall in the right ventricular end-dia-stolic and pulmonary arterial pressures, while inpatient D. K., there were no changes in thesepressures.

The two patients (W. H., W. Da.) in thisgroup with congestive failure had appreciable de-creases in cardiac output after phlebotomy withno changes in the right ventricular end-diastolicpressure. Patient W. H., showed a moderate

L IE

RIGT ATIL

ARTKOO

PRI

M%

i ...ii .... ....

_ A . ........Ez_ __A

>

-1-

619

.Xli1isil 1; 11 '; li 1' 1p:, .i

W. JUDSON, W. HOLLANDER, J. HATCHER, M. HALPERIN, AND I. FRIEDMAN

decrease in the pulmonary and brachial arterialpressures. In contrast, patient W. Da. had no

fall in the pulmonary and brachial arterial pres-sures even in the presence of a marked reductionin cardiac output and hematocrit. There was a

very slight tendency for the arterial oxygen satu-ration to increase after phlebotomy (+ 1.7 per

cent), but the change lacked statistical signifi-cance (P = 0.10). The oxygen consumption,hematocrit, and pulse rate did not change appre-

ciably in the group as a whole.

DISCUSSION

Normal subjects, patients with compensated hy-pertensive cardiovascular disease, and patientswith cor pulmonale with or without congestivefailure all appeared to have a similar response incardiac output after venesection or venous con-

gestion of the limbs, procedures which are believedto produce a reduction in the "effective blood vol-ume." This response was characterized by a de-crease in cardiac output which was usually notassociated with measurable changes in the rightventricular end-diastolic pressure. In contrast,patients with congestive heart failure due to hy-pertension, coronary, or valvular heart disease hadusually a slight increase, occasionally no change,or rarely a slight decrease in cardiac output.Again these changes in cardiac output were ob-served in individual cases to be independent ofchanges in the right ventricular end-diastolic pres-

sure, which either remained the same or decreased.The group as a whole, however, showed a signifi-cant reduction in the right ventricular diastolicpressure.

Collateral studies in this laboratory have shownthat patients with different types of congestivefailure also may show characteristic changes incardiac output in response either to exercise (6)or to the acute infusion of saline solution (7).Thus, in response to these stimuli, normal sub-jects, patients with compensated heart disease, andpatients with cor pulmonale in failure usually in-crease their cardiac outputs. Patients with hyper-tension, coronary artery, and valvular heart dis-ease with congestive failure have relatively fixedcardiac outputs which do not increase normallyafter exercise or saline infusion. Furthermore,as shown in the present study, the cardiac output

in such patients usually does not decrease nor-mally after stimuli which are believed to causereductions in the "effective blood volume." Theobservations suggest that the heart in hyperten-sive heart disease without congestive failure andin cor pulmonale without or even with congestivefailure may have greater reserve and may be moreresponsive to changes in right heart filling, whilein the patients with so-called "low output failure"the heart is usually less responsive to changes inright heart filling.

The responses in cardiac output in the presentstudy are similar to those previously reported (1).However, increases in cardiac output in patientswith "low output" congestive heart failure wereobserved less frequently, and when observed,were smaller in magnitude than those found byHowarth, McMichael, and Sharpey-Schafer (3).An increase in cardiac output has been reportedto occur occasionally after phlebotomy in pa-tients with chronic cor pulmonale and severe con-gestive heart failure (8). By contrast, in our twocases of chronic pulmonary disease with severecongestive failure the cardiac output fell afterphlebotomy. As just mentioned, in a larger seriesof patients with cor pulmonale and congestivefailure, a normal increase in cardiac output wasfrequently found in response to exercise or to anacute infusion of hypertonic saline solution (9).

Warren, Brannon, Stead, and Merrill (2) foundthat the response of cardiac output depends notsolely upon changes in the right atrial pressure.McMichael, and his co-workers (1, 3, 10) havealso observed falls in the right atrial pressureafter reduction in "effective blood volume."Measurements of mean right atrial pressure, how-ever, may not accurately reflect right ventricularfilling pressure. The measurements of the rightventricular end-diastolic pressure obtained in thepresent study did not show a uniform decreaseafter venesection or trapping of blood in the ex-tremities by venous congestion in patients with"low output" type of failure. Furthermore, theresponse in cardiac output was found to be in-dependent of the changes in the right ventricularend-diastolic pressure. These results, however,do not necessarily bear upon the relevancy ofStarling's Law to the intact human heart sinceneither the right ventricular end-diastolic volumenor the pericardial pressure was measured. The

.620

CARDIAC RESPONSESTO VENOUSCONGESTIONAND PHLEBOTOMY

data suggest, however, that factors other thanright heart filling are also important in determin-ing cardiac output. These factors might, andundoubtedly do, include some or all of the follow-ing: 1) the degree or "stage" of congestive fail-ure and the associated state of myocardial metabo-lism. This factor would include those considera-tions ordinarily implied by the terms "myocardialreserve," "cardiac fatigue," and "cardiac tone";2) the involvement in the failure of one or morechambers of the heart. Thus, failure of the leftand right ventricles might well behave differentlythan failure of either one alone; 3) the amount ofpulmonary vascular and even of peripheral vascu-lar disturbances associated with the failure; 4)the amount of mechanical defect such as valvularincompetence associated with the failure.

The main factors controlling pulmonary ar-terial pressure (in relation to intrapleural pres-sure) are cardiac output, pulmonary arteriolarresistance, and resistance to flow through the leftside of the heart. Since the latter two measure-ments were not obtained in this study they cannotbe evaluated. However, it can be stated that thereductions in cardiac output observed in normalsubjects and in patients with cor pulmonale werenot consistently associated with falls in pulmonaryarterial pressure. In patients with left ventricu-lar failure who had rises in cardiac output afterphlebotomy the reduction in pulmonary arterialpressure was attributed partly to a decrease inresistance to flow because of a lower left atrialpressure. However, since "wedge" or "pulmo-nary capillary" pressures were not taken becauseof the already great complexity of the procedure,no definite observations bearing on this pointare available.

CONCLUSIONS

During venesection or venous congestion ofthe limbs (acute reductions of "effective bloodvolume"),

1. Normal subjects, patients with compensatedcardiovascular disease, and patients with cor pul-monale with or without failure had a fall in cardiacoutput.

2. Patients with congestive heart failure due tovalvular, hypertensive, or coronary heart diseaseusually had a slight rise, occasionally no change,or rarely a slight fall in cardiac output.

3. The response in cardiac output was notnecessarily accompanied by a change in the rightventricular end-diastolic pressure.

4. A reduction in the pulmonary arterial pres-sure was not consistently observed. Decreases incardiac output in some patients may explain thefall in pulmonary arterial pressure.

ACKNOWLEDGMENTThe authors gratefully acknowledge the technical as-

sistance of Miss Janice McMorrow, Miss Adele Rymut,and Miss Margaret Sullivan.

REFERENCES

1. McMichael, J., and Sharpey-Schafer, E. P., Cardiacoutput in man by a direct Fick method. Effects ofposture, venous pressure change, atropine, andadrenaline. Brit. Heart J., 1944, 6, 33.

2. Warren, J. V., Brannon, E. S., Stead, E. A., Jr., andMerrill, A. J., The effect of venesection and thepooling of blood in the extremities on the atrialpressure and cardiac output in normal subjects withobservations on acute circulatory collapse in threeinstances. J. Clin. Invest., 1945, 24, 337.

3. Howarth, S., McMichael, J., and Sharpey-Schafer,E. P., Effects of venesection in low output heartfailure. Clin. Sc., 1946, 6, 41.

4. Harvey, R. M., Ferr6r, M. I., Cathcart, R. T., Rich-ards, D. W., Jr., and Cournand, A., Some effectsof digoxin upon the heart and circulation in man.Digoxin in left ventricular failure. Am. J. Med.,1949, 7, 439.

6. Gorlin, R., Haynes, F. W., Goodale, W. T., Sawyer,C. G., Dow, J. W., and Dexter, L., Studies of thecirculatory dynamics in mitral stenosis. II. Al-tered dynamics at rest. Am. Heart J., 1951, 41,30.

6. Haldane, J. S., Methods of Air Analysis. London,Griffin, 1912.

7. Van Slyke, D. D., and Neill, J. M., The determina-tion of gases in blood and other solutions by vac-uum extraction and manometric measurementI. J. Biol. Chem., 1924, 61, 523.

8. Judson, W. E., Hollander, W., Hatcher, J. D., Hal-perin, M. H., and Friedman, I., The effects ofexercise in the supine position on the cardio-renalhemodynamics and excretion of electrolytes andwater. To be published.

9. Hollander, W., Judson, W. E., and Friedman, I.,Cardiovascular responses to hypertonic saline inpatients with and without congestive heart failure.To be published.

10. Howarth, S., McMichael, J., and Sharpey-Schafer,E. P., Effects of oxygen, venesection and digitalisin chronic heart failure from disease of the lungs.Chin. Sc., 1946, 6, 187.

621