Bga Pada Infark

and THOMAS KILLIP IIISIDNEY J. FILLMORE, ARMENIO C. GUIMARÃES, STEPHEN S. SCHEIDT

Myocardial InfarctionBlood-Gas Changes and Pulmonary Hemodynamics following Acute

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1972 American Heart Association, Inc. All rights reserved.

75231is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TXCirculation

doi: 10.1161/01.CIR.45.3.5831972;45:583-591Circulation.

http://circ.ahajournals.org/content/45/3/583located on the World Wide Web at:

The online version of this article, along with updated information and services, is

http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer

of the Web page under Services. Further information about this process is available in thewhich permission is being requested is located, click Request Permissions in the middle columnClearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of the CopyrightCirculationoriginally published in Requests for permissions to reproduce figures, tables, or portions of articlesPermissions:

by guest on December 19, 2013http://circ.ahajournals.org/Downloaded from by guest on December 19, 2013http://circ.ahajournals.org/Downloaded from

http://circ.ahajournals.org/content/45/3/583

http://circ.ahajournals.org/content/45/3/583

http://www.ahajournals.org/site/rights/

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://www.lww.com/reprints



http://circ.ahajournals.org/




Blood-Gas Changes and PulmonaryHemodynamics following Acute

Myocardial InfarctionBy SIDNEY J. FILLMORE, M.D., ARMENIO C. GUIMARXES, M.D.,

STEPHEN S. SCHEIDT, M.D., AND THOMAS KILLIP III, M.D.

SUMMARYArterial and mixed venous oxygen tensions were measured in 24 patients following

acute myocardial infarction while they were breathing air and 100% oxygen. Totalvenous admixture and the right-to-left shunt during 100% oxygen breathing were cal-culated. These data were related to the pulmonary arterial diastolic pressure, thecardiac index, and the central blood volume.

Patients with myocardial infarction that was not complicated by congestive failurehad blood gases, pulmonary shunts, and pulmonary arterial diastolic pressures com-

parable to control patients who were at rest in bed.When congestive failure complicated myocardial infarction, arterial blood oxygen

tension was lower, pulmonary shunting was increased, and the pulmonary arterialdiastolic pressure was elevated. Cardiac index and central blood volume were usuallynormal.The present data quantitate the contribution of anatomic shunting to the hypoxemia

observed in myocardial infarction. Hypoxemia and increased anatomic shunting are

closely correlated to the degree of elevation of pulmonary arterial diastolic pressure.The interrelationships of arterial hypoxemia, venous admixture, arterial-alveolar oxygengradient, and pulmonary arterial diastolic pressure suggest that pulmonary venouscongestion is an important determinant of the hypoxemia and shunting observed inpatients with acute myocardial infarction.

Additional Indexing Words:Pulmonary arterial diastolic pressureRight-to-left shunt in lung Oxygen breathing

IN THE PAST 5 years, arterial hypoxemia abnorhas been recognized as a common occur- heart

rence during the first few days of hospitaliza- normction after acute myocardial infarction.1-10 nismsRecently, the magnitude of the hypoxemia has cardiabeen related to the clinical condition of the in apatient.10 Arterial oxygen tension is most studie

From the Division of Cardiology, Department ofMedicine, Cornell University Medical College, NewYork, New York.

Supported in part by the Myocardial InfarctionResearch Unit Contract 43-67-1439 and the Cardio-vascular Training Grant 5-T12-HE05789-03, in partby the Commonwealth Fund (Dr. Guimaraes), and inpart by the New York Heart Association (Dr.Fillmore was a Clinical Science Fellow).Circulation, Volume XLV, March 1972

pulmdludecapill

Pulmonary vascular congestion

rmal in patients with clinical evidence offailure or shock and returns toward

al as the condition improves. Mecha-;postulated for the hypoxemia of myo-al infarction have been lucidly discussedrecent symposium." Several previous

es have attempted to assess the role ofonary shunts in this hypoxemia.'6' 9 In-d in our concept of anatomic shunts arelaries which are completely unventilated

Address for reprints: Dr. Sidney Fillmore, MaricopaCounty General Hospital, 2601 E. Roosevelt, Phoenix,Arizona 85008.

Received April 7, 1971; revision accepted forpublication October 22, 1971.

583

by guest on December 19, 2013http://circ.ahajournals.org/Downloaded from



FILLMORE ET AL.

(e.g., fluid-filled or collapsed), anatomic vas-cular communications, bronchial venous flow,thebesian venous drainage, and collapsed ter-minal airways. The extent to which shuntingcontributes to the total arterial unsaturationhas been estimated indirectly from arterial ox-ygen mixtures.'--' "Anatomic shunting" maybe more precisely measured by administrationof 100% oxygen and analysis of mixed venousand systemic arterial blood samples.12 1. How-ever, in previous studies 100% oxygen has notalways been utilized and samples of venousblood from the pulmonary artery were notoften obtained. Thus, reported shunt calcula-tions have usually been based on reasonableestimates rather than actual measurements.Furthermore, the blood-gas measurements andcalculations of shunt flow have not beencorrelated with hemodynamic measurements.

In the present study we quantitated themagnitude of right-to-left "anatomic shunting"of blood through the lungs in a group ofpatients with acute myocardial infarction.Both arterial and mixed venous oxygentensions were measured before and after thebreathing of 100% oxygen. The results havebeen related to the clinical findings andcertain hemodynamic parameters, includingthe pulmonary arterial diastolic pressure, thecardiac index, and the central blood volume,in an attempt to elucidate some of theinterrelationships of pulmonary and cardio-vascular derangements which may accompanyacute myocardial infarction.

MethodsBlood-gas and hemodynamic data were

obtained from 24 patients with definite acutemyocardial infarction within 24 hours of theiradmission to the Myocardial Infarction ResearchUnit of the New York Hospital. The diagnosis ofacute myocardial infarction was establishedaccording to criteria previously published.14 Theages of the patients ranged from 44 to 79 years(mean, 62.7 years). Nineteen were men. In nineof the patients, studies were done on successivedays. Five control studies were obtained in fourpatients hospitalized for investigation of sponta-neous angina pectoris. None of these four hadclinical evidence of congestive heart failure.Patients on whom the diagnosis of infarction wasconsidered but not proved were not included in

either group. None of the patients was grosslyobese. All 28 subjects were supine in bed withone or two pillows for comfort during theprocedure. Some had received analgesics within 4hours of our sampling, but all were fully alert andcooperative. No relation was apparent betweenthe level of sedation and the presence of heartfailure.The various maneuvers were carried out in a

calm and efficient manner, and the subjectsappeared to be in a steady state. The nature andpurpose of the study was explained to eachpatient prior to initiation of the procedures, andinformed consent was obtained. The principles ofthe Declaration of Helsinki were strictly followed.An indwelling nylon arterial catheter (i.d., 1.0

mm; o.d., 1.4 mm) was inserted via the Seldingertechnic. In 19 patients a similar catheter wasadvanced from an antecubital vein into thepulmonary artery while the intravascular pressurepulse was monitored to substantiate the locationof the catheter tip. After insertion of the catheter,the patient relaxed for at least 20 min in order toreturn to a steady state prior to measurements.Patients were asked to breathe quietly anid torefrain from taking deep breaths so that themeasurements would better reflect the actualclinical state.

Arterial and mixed venous blood samples weredrawn into heparinized plastic syringes while thepatient was breathing room air. After administra-tion of 100% oxygen for 20 min, additionalsamples were obtained. The oxygen was suppliedthrough a mouthpiece and a nonrebreathingvalve while the nose was occluded. A 10-inchlength of wide-bore rubber tubing was attachedto the exhaust limb of the valve to preventbackflow of air through the valve during theinitial moments of inspiration.

Blood samples were iced and taken immediate-ly to the laboratory where they were analyzed foroxygen tension, carbon dioxide tension, and pHwithin 10 min of collection. Determinations wereperformed with a Radiometer pH meter no. 27with a gas monitor. The arterial (PaO,) andmixed venous (Pmv(-,) oxygen tensions weremeasured with an oxygen electrode calibratedaccording to the anticipated oxygen tension: formixed venous samples a gas mixture with knownoxygen tension in the range of 30 to 35 mm Hgxvas used; for arterial samples during airbreathing, the known mixture used was in therange of 80 to 85 mm Hg. For arterial samplesobtained during oxygen breathing, calibrationwas accomplished with humidified oxygen. Selec-tion of the calibrating gas according to theanticipated oxygen tension diminishes the prob-lem of alinearity occasionally encountered withthe polarographic method.

Circulation, Volume XLV, March 1972

584




PULMONARY HEMODYNAMICS AFTER AMI

Ten 5-ml samples of blood exposed to 100%oxygen in a rotating tonometer for more than 30min and handled exactly as the patients' speci-mens yielded oxygen tension of 611 + 21 mm Hg.This result is 102 mm Hg lower than predicted. Avalue lower than that theoretically postulated forblood oxygen tension when wet gas calibration isused has been found and this finding has beendiscussed by others.15 16 The discrepancy isprobably due to errors in both handling andmeasuring samples with high 02 tensions. Wehave not attempted to correct for these variablemethodologic errors in calculating our results. Anapproximate correction factor would have re-

duced the calculated anatomic shunt by less than1% and the alveolar-arterial 02 difference (A-aD02) by approximately 60 to 100 mm Hg.The A-aDO was calculated by subtracting the

arterial oxygen tension measured during oxygenbreathing from the theoretical alveolar oxygentension. Alveolar oxygen tension was assumed tobe equal to the barometric pressure less thecombined tensions of alveolar CO2 (presumedequal to the measured arterial Pcoj) and watervapor at 37.8°C.Oxyhemoglobin saturation was obtained from a

Severinghaus blood-gas calculator* using themeasured values for oxygen tension and pH.From the oxygen tension, saturation, and hemo-globin concentration (cyanmethemoglobin meth-od), the total venous admixture while the patientwas breathing room air was calculated accordingto the formula:

Cc-Ca QsCc-Cmv Q

Cc represents the pulmonary end-capillary bloodoxygen content. Alveolar oxygen tension was

assumed to be 104 mm Hg. Ca and Cmv representthe respective arterial and mixed venous bloodoxygen contents. Qs represents the portion of thetotal cardiac output (Q) that is shunt flow. Theanatomic shunt was calculated in the same

manner, using the results from the correspondingsamples obtained during oxygen breathing.

Pulmonary arterial diastolic pressure (PADP)and cardiac output were measured during quietair breathing a few minutes subsequent to theoxygen studies. Pressures were measured with a

Statham 23 Db transducer. Zero reference levelwas 10 cm below the sternal angle as determinedwith a carpenter's spirit level. Cardiac output(CO) was measured by the indicator-dilutionmethod,17 using indocyanine green as an indicator,a Waters X302 Densitometer, and an Electronics

for Medicine photographic recorder. Indicator-dilution curves were integrated by manual meth-ods after replotting on semilog paper. The mean ofthree separate curves was accepted as the output.Central blood volume (cbv), the content of theintravascular space between the pulmonary arte-rial injection site and the sampling site in theproximal aorta, was calculated from the dyecurves by the formula:

arrival time + mean transit timecbv-= 60 x CO.

The results were expressed in relation to the bodysurface area.18

Results

Blood-Gas Findings

Results obtained from analysis of bloodgases are listed in table 1. Mean arterialoxygen tension while the patient was breath-ing air was 78 12 (SD) mm Hg in theabsence of heart failure but fell to 59 + 14 mmHg (P > 0.001), when failure was detectedclinically. The decrease in arterial oxygentension associated with heart failure resultedin a rise in the mean venous admixture(calculated from the data obtained while thepatient breathed air) from 12.2-+±7.8% with-out congestive failure to 21.2 10.5% (P>0.05) with it.

Breathing oxygen for 20 min raised the Pa02to 528 ±55 mm Hg in patients withoutfailure. The response to oxygen breathingwas significantly lower, 376 104 mm Hg(P> 0.001), when failure was present. As a

result, calculated A-aDO, was significantlyincreased in the presence of heart failure,averaging 317 + 115 mm Hg in the patientswith congestion and 138 58 in those withoutit (P > 0.001). In the patients with pulmonaryedema, Pa02 decreased still further to221 105 mm Hg. The abnormal response to02 breathing associated with heart failureresulted in a significant increase in thecalculated anatomic shunt. Thus, the anatomicshunt rose from 5.4 + 2.9% of the cardiacoutput in patients at rest in bed withoutfailure to 9.2 3.7% when failure was present(P > 0.02).The calculated venous admixture is greater

than the anatomic shunt. The difference*The London Company, Westlake, Ohio.Circulation, Volume XLV, March 1972

585




FILLMORE ET AL.

Table 1

Mean Values for^ Blood-Gas Analyses and Calculationts

Myocardial infarctionPulmonary

No myocardial infarction No CHF* CHF, mild edema

No. of observations . 12 18 3Pao2

Air(mm Hg) 7,5 5 78 - 12 39 - 14t 57 3100% 02 (mm Hg) 317 -51 528 -35 376 104t 221 105

Alveolar-arterial 174 52 138 - 58 317 - l1t 443 - 88gradient (mm Hg)

Anatomic shunt (%c) 6.3- 1.9 , 5.4- 2.9 9.2 3.7tVenous admixture (Co 10 1.1 12.2 - 7.8 21.2 - 10.5§Venous admixture 3.0 0.5) 6.8 - 5 13.1 - 6.5$minus anatomicshunt (net %)

*Values in patients without congestive failure are Inot significantly different from the controlgroup.tP < 0.001 compared to the patients without failure.tP < 0.02 compared to the patients without failure.§P < 0.05 compared to the patients without failure.Abbreviation: CHF = coigestive heart failure.

between these two calculations represents thecombined contribution of ventilation-perfu-sion imbalances and of possible diffusiondisturbances to the hypoxemia, and like theanatomic shunt is significantly larger, 13.1 ±6.5% when failure is present compared to6.8 + 5% without failure (P > 0.02).

Data obtained from the five control studiesin patients who had coronary artery diseasebut no evidence of acute infarction or heartfailure were not significantly different fromthose obtained in the 12 patients with acutemyocardial infarction and no heart failure.Mean values for PacO2 during air breathing

was 39.1 ±3 mm Hg in patients with acuteinfarction without heart failure compared to40.3 + 3 mm Hg in those with congestivefailure. Mean arterial pH was also similar inthe two groups, averaging 7.42 + 0.03 with-out, and 7.41 + 0.04 with, heart failure.Carbon dioxide tensions in both groups duringoxygen breathing were within ± 2 mm Hg ofthe values during air breathing and attest tothe steady state of the patients during thepresent study.

Hemodynamic Findings

The hemodynamic data are recorded intable 2. Mean pulmonary arterial diastolic

pressure of 8.8 ± 3.6 mm Hg in patientswithout cardiac failure did not differ signifi-cantly from the value observed in patientswithout infarction. Clinical evidence of heartfailure was associated with a significantlyhigher pulmonary arterial diastolic pressure,16.6 ± 7.0 mm Hg (P > 0.005). There was nosignificant difference in pulmonary arterialpulse pressure. The two patients with thehighest pulmonary arterial diastolic pressuresboth went into cardiogenic shock several daysafter the studies and eventually died.The cardiac index was nearly identical in

patients with and without congestive failure.Calculated central blood volume was signifi-cantly higher in patients with failure,709 + 186 ml, compared to 560 ± 119 in thosewithout cardiac failure (P > 0.05).

Hemodynamic and Blood-Gas RelationshipsArterial oxygen tension, measured while the

subjects were breathing room air variedinversely with the pulmonary arterial diastolicpressure. The data for all subjects are plottedin figure 1; the coefficient of correlation isr -0.78. Although the data are analyzed onthe basis of a postulated linear relationship,the plot suggests the possibility of a two-component curve with a break at a pulmonary


586




PULMONARY HEMODYNAMICS AFTER AMI

Table 2

Mean Values for Hemnodynamic Data

No myocardial Myocardial infarctioninfarction No CHF CHF

No. of observations 3 8 14Pulmonary arterial diastolic pressuire 7.3 1.7 8.8 3.6* 16.6 7.0*(mm Hg)

Pulmonary arterial pulse pressure (mm Hg) 9.0 0.8 13.1 3.7 13.7 3.4Cardiac index (liters/min/M2) 2.0.5 0.05t 2.76 - 0.52t 2.70 0.67Central blood volume (ml/M2) .520 - 2 560 - 119t 709 186t

*Difference between means significant, P < 0.005.tDifference between means significant, P < 0.01.tDifference between means significant, P < 0.0.5.Abbreviation: CHF = congestive heart failure.

0 80I

E

E 70z0V)Z 60LU

z

? 50 _X0

' 40

< 30

0

0A A0

0

A

NoMI A

MI, no CHF 0

MI, CHF 0

0

.00

r = -0.78

p < .005

0

0

0

0

10 15 20 25PULMONARY ARTERIAL

DIASTOLIC PRESSURE. mm Hg

0

C3)I

E

E

0LI)

zLUJzLJ0

X0

30 35 U-

Figure 1

Arterial oxygen tension during the breathing of airplotted in relation to pulmonary arterial diastolic pres-

sure. p refers to the significance of the correlation co-

efficient r.

arterial diastolic pressure of 15 mm Hg. Asimilar inverse relationship was observedbetween the arterial oxygen tension achievedduring inhalation of 100% oxygen and thepulmonary arterial diastolic pressure, r

- 0.85 (fig. 2). The calculated anatomic shuntvaried directly with the pulmonary arterialdiastolic pressure (fig. 3). The largest shuntswere recorded in patients with the highest pul-monary arterial diastolic pressure (r = 0.71).The A-aD02 (r = 0.80) and the venous admix-ture (r = 0.72) also were significantly cor-

related with the pulmonary arterial diastolicCirculation, Volume XLV, March 1972

5 10 15 20 25 30

PULMONARY ARTERIALDIASTOLIC PRESSURE, mm Hg

Figure 2

Arterial oxygeni tension during the breathinig of oxy-

gen plotted in relation to the pulmonary arterialdiastolic pressure.

pressure. Neither the cardiac index nor thecentral blood volume could be significantlyrelated to pulmonary arterial pressures.

When congestive heart failure was present,the arterial oxygen tensions achieved duringboth air and oxygen breathing were lower,while the venous admixture, the anatomicshunt, the alveolar-arterial oxygen gradient,and the pulmonary arterial pressures were

higher than in the patients who were not infailure (fig. 4).

587

c




FILLMORE ET AL.

0 0

0

0

A

0

0

0

NoMI

MI, no CHF

MI, CHF0

A

0

0

r -0.71

p < .001

5 10 15 20 25 30 35PULMONARY ARTERIAL

DIASTOLIC PRESSURE, mm HgFigure 3

Anatomic shunt plotted in relation to the pulmonaryarterial diastolic pressure.

Serial ObservationsFigure 5 illustrates the serial changes in a

patient who was initially studied duringcardiogenic shock and gradually improved torecovery on the fourth hospital day. At thetime of the initial study, arterial Pa02 was 43mm Hg during air breathing and 102 mm Hg

15

c0)

-Ua)

;_ 10zD

U

z

0

0

0

0

0

-U-0

0

00

NoMI

0

0

0

.0

@0*0

@00

0

@0

.0

0

MI, no CHF MI,CHF

during administration of 100% oxygen. As thepatient improved clinically, the A-aDo0 grad-ually fell and the value of Pao2 rose.

DiscussionWork from this laboratory10 and by

others'-9 has established that hypoxemia is acommon accompaniment of acute myocardialinfarction. Four mechanisms for this hypoxe-mia have been postulated. These include: (1)decreased alveolar ventilation, (2) ventilation-perfusion imbalances, (3) diffusion abnormal-ities, and (4) increased pulmonary arteriove-nous shunting.

Since arterial carbon dioxide tension isalmost always normal or diminished followingacute myocardial infarction, -4 10 it is likelythat overall alveolar hypoventilation is not asignificant contributor to the hypoxemia,except perhaps in severe cardiogenic shock."

It has been shown that the ratio of deadspace to tidal volume is elevated in mostpatients with acute myocardial infarction, and

600

500

E

LI)LI'

u-Li

a~

400 h

300_

200h

100o

0'

TIMECLINICALSTATE

AIR BREATHING

11A.M. 3PM. 10A.M.DAY ONE DAY TWO

Cordiog.nc Pulmonary MildShock Edema CHF

11AM.DAY FOURNo CHF

Figure 4

Anatomic shunt distributed in accord with the clinicalfindings of congestive failure. Although there is con-

siderable scatter, note the definite trend toward in-creased shunting in the presence of congestive failure.Horizontal lines indicate mean values for each group.

MI myocardial infarction; CHF= congestive heartfailure.

Figure 5Serial oxygen tensions during progressive recovery

from cardiac shock. Plotted are the arterial oxygen

tensions, obtained while the patient breathed room

air and then 100% oxygen, and the alveolar-arterialgradient. Arterial oxygen tensions rose and the alveo-lar-arterial gradient decreased as left ventricular failurediminished.


588

cQ)U

1-(L)O-

zDILI)U

0

z a a B^ . ^ ^ * ^ k . . . - - - - - - - by guest on December 19, 2013http://circ.ahajournals.org/Downloaded from



PULMONARY HEMODYNAMICS AFTER AMI8

that the physiologic dead space is usuallyincreased.1-3 Thus, some portion of theobserved hypoxemia is attributable to ventila-tion-perfusion imbalances.The presence and importance of diffusion

abnormalities remain unsettled. Hardy andassociates4 reported that steady-state carbonmonoxide diffusing capacity was normal fol-lowing myocardial infarction, even in thepresence of congestive failure. On the otherhand, Valencia and Burgess6 used the single-breath carbon monoxide technic and reportedthat, when heart failure was not evident clin-ically, the pulmonary diffusing capacity(D,,(,,) ranged from 51 to 122% of predictedvalue. With congestive failure, DLcO was be-tween 67 and 100% of the predicted value fortheir laboratory. The reasons for these dis-pirate results are not apparent and awaitclarification.The contribution of intrapulmonary shunt-

ing to the total venous admixture may bequantitated by administering 100% oxygen.'2 13After a suitable period (usually 10 min issufficient in the absence of intrinsic lungdisease) nitrogen is no longer present inthe alveoli, and even the most inadequatelyventilated lung units contain 0.,, C02, andH2O vapor. Sampling and analysis of arterialand mixed venous blood allow calculation ofthe percentage of the blood flow which isshunted from the pulmonary to the arterialcirculation without undergoing gas exchange.It is assumed that at this high alveolar oxygentension there is no appreciable diffusionbarrier."' 19The present data clearly indicate that

anatomic shunting as measured by the admin-istration of 100% oxygen contributes to thehypoxemia observed in acute myocardialinfarction. Furthermore, this shunting is in-creased when congestive failure and pulmo-nary edema develop (figs. 3 and 4). Thesefindings are in accord with the results ofHiggs,3 but they are at variance with those ofSukumalchantra and associates,9 who foundno difference in the magnitude of shunting inpatients with uncomplicated myocardial in-farction, those with left ventricular failure,Circulation, Volume XLV, March 1972

and those in shock. The reasons for thesedivergent results are not apparent, but may bedue in part to the fact that most of thesubjects in the latter study did not receive100% oxygen and that the samples of mixedvenous blood were obtained from the venacava or right atrium rather than the pulmo-nary artery. While it is often claimed that theformer sampling sites adequately representmixed venous blood, this has not beendocumented in acutely ill patients, especiallywhen shock is present.The method we have employed to deter-

mine the magnitude of shunting within thelung does not define the anatomic site(s) ofthe shuntlike effect. Possibilities include bloodperfusing the capillaries of completely unven-tilated (e.g., fluid-filled) or atelectatic alveoli,anatomic vascular communications, bronchialvenous flow, and thebesian venous drainage.We did not measure the effects of deepbreathing. This maneuver opens temporarilycollapsed terminal airways and would almostcertainly have transiently altered arterialblood oxygen. Collapsed airways most likelycontribute to the anatomic shunt in the patientwith myocardial infarction. However, weattempted to maintain a uniform steady stateand did not induce deep breathing.The analytical errors inherent in collecting

and measuring blood samples containing highoxygen tensions total approximately 100 mmHg. This increases the numerator of the shuntequation by approximately 0.3 ml 02/100 mlof blood and leads to a small (less than 1%)overestimate of the shuntlike effect. Theresults do not materially affect the interclassdifferences that we are reporting. Arterialoxygen tensions achieved in our patientsduring oxygen breathing are comparable tothose reported by others.1-6Pulmonary arterial end-diastolic pressure is

closely correlated with a commonly acceptedindicator of left ventricular failure, the leftventricular end-diastolic pressure.20 It is some-what lower than left ventricular end-diastolicpressure measured directly, especially whenthe latter is greatly increased. In the absenceof pulmonary vascular disease, however,

589




FILLMORE ET AL.

elevated pulmonary artery diastolic pressurereflects left ventricular dysfunction. Correla-tion between mean pulmonary arterial wedgepressure and pulmonary arterial diastolicpressure in patients with acute myocardialinfarction has been noted by others.21 In ourpatients the pulmonary arterial pulse pressurewas not increased, and it is not likely thatchanges in pulmonary vascular resistanceinfluenced the relationship between left ven-tricular and pulmonary arterial pressures.The correlations between arterial oxygen

tensions and pulmonary arterial diastolicpressures both when the subject was breathingair (fig. 1) and oxygen (fig. 2) suggest to usthat gas exchange is influenced to a consider-able degree by pulmonary venous pressure,even in the absence of clinically recognizablecongestive heart failure. We believe that thisvascular congestion and the concomitantdefects in gas exchange occur as a continuousand interrelated distribution in acute myocar-dial infarction. A similar correlation, betweenthe arterial oxygen tension and the pulmonaryarterial (capillary) wedge pressure, has re-cently been noted by another group.21While the blood-gas measurements were

well correlated with the elevation of pulmo-nary arterial diastolic pressure, cardiac outputand central blood volume were poorly corre-lated with this indicator of ventricular func-tion. Cardiac output depends upon manyfactors, including metabolic demand, preload,contractility, and afterload. Preload, orventricular filling pressure, influences pulmo-nary function, but it is only one of the manyfactors regulating cardiac performance. It isnot surprising, therefore, that the level ofblood flow cannot be related to arterialoxygenation.

AcknowledgmentsWe gratefully acknowledge the technical assistance

of Miss Jacqueline Drangel, Mrs. Nora Kimball, andMr. Stuart Gordon. Dr. William Briscoe gavegenerously of his time and advice. Miss ElizabethJohnson and Mrs. Cheryl Shallow provided secretarialassistance.

References1. McNICOL MW, KIRBY BJ, BHOOLA KD, EVEREST

ME, PRICE HV, FREEDMAN SF: Pulmonaryfunction in acute myocardial infarction. BritMed J 2: 1270, 1965

2. PAIN MCF, STANNARD M, SLOMAN G: Distur-bances of pulmonary function after acutemyocardial infarction. Brit Med J 2: 591,1967

3. HIGGS B: Factors influencing pulmonary gasexchange during the acute states of myocardialinfarction. Clin Sci 35: 115, 1968

4. HARDY WE, AYRES SM, KEYLOUN V, GRACE WJ:Causes of hypoxemia and alkalemia in acutemyocardial infarction. Clin Res 16: 370,1968

5. STORSTEIN 0, RASMUSSEN K: The causes ofarterial hypoxemia in acute myocardial infarc-tion. Acta Med Scand 183: 193, 1968

6. VALENCIA A, BURGESS JH: Arterial hypoxemiafollowing acute myocardial infarction. Circula-tion 40: 641, 1969

7. FOSTER GL, CASTEN GG, REEVES TJ, HURST DC:The effects of oxygen breathing in patientswith acute myocardial infarction. CardiovascRes 3: 179, 1969

8. SUKUMALCHANTRA Y, LEVY S, DANZIG R, RUBINSS, ALPERN H, SWAN HJC: Correcting arterialhypoxemia by oxygen therapy in patients withacute myocardial infarction. Amer J Cardiol24: 838, 1969

9. SUKUMALCHANTRA Y, DANZIG R, LEVY SE, SWANHJC: The mechanism of arterial hypoxemia inacute myocardial infarction. Circulation 41:641, 1970

10. FILLMORE SJ, SHAPIRO M, KILLIP T: Arterialoxygen tension in acute myocardial infarction:Serial analysis of clinical state and blood gaschanges. Amer Heart J 79: 620, 1970

11. AYRES SM, MUELLER H, GIANNELLI S, FLEMIINGP, GRACE WJ: The lung in shock. Amer JCardiol 26: 588, 1970

12. BERGGREN SM: The oxygen deficit of arterialblood caused by non-ventilating parts of thelung. Acta Physiol Scand 4 (suppl II): 19,1943

13. FINLEY TN, LENFANT C, HAAB P, PIIPER J,RAHN H: Venous admixture in the pulmonarycirculation of anesthetized dogs. J Appl Physiol15: 418, 1960

14. KILLIP T, KIMBALL J: Treatment of myocardialinfarction in a coronary care unit. Amer JCardiol 20: 457, 1967

15. RHODES PC, MOSER KM: Sources of error inoxygen tension measurement. J Appl Physiol21: 729, 1966

16. MORAN F, KETTEL LJ, CUGELL DW: Measure-ment of blood PO., with the microcathodeelectrode. J Appl Physiol 21: 725, 1966


590




PULMONARY HEMODYNAMICS AFTER AMI5

17. STEWART GN: Researches on circulation time andon influences which affect it: IV. Output of theheart. J Physiol 22: 159, 1897

18. KEYS JR, HETZEL PS, WOOD EH: Revisedequations for calculation of blood flow andcentral blood volume from indicator dilutioncurves. J Appl Physiol 11: 385, 1957

19. LILIENTHAL JL JR, RILEY RL, PROENIMEL DD,FRANKE RE: An experimental analysis in manof the oxygen pressure gradient from alveolarair to arterial blood during rest and exercise at

sea level and at altitude. Amer J Physiol 147:199, 1946

20. BOUCHARD RJ, GAULT JH, Ross J JR: Comparisonof pulmonary arterial end-diastolic pressures inpatients with and without left ventriculardisease. Circulation 39 (suppl III): III-49,1969

21. LASSERS BW, GEORGE M, ANDERTON JL,HIGGINS MR, PHILP T: Left ventricularfailure in acute myocardial infarction. Amer JCardiol 25: 511, 1970


591




Documents

Bga Pada Infark