Assessment byradionuclide angiographyofright emphysema

7
Thorax 1983;38:494-500 Assessment by radionuclide angiography of right and left ventricular function in chronic bronchitis and emphysema W MACNEE, QF XUE, WJ HANNAN, DC FLENLEY, CJ ADIE, AL MUIR From the Departments of Medicine, Respiratory Medicine, and Medical Physics and Medical Engineering, University of Edinburgh ABSTRACT Non-invasive measurements of right and left ventricular ejection fraction (RVEF, LVEF) by multiple-gated equilibrium radionuclide ventriculography were performed in 18 con- trol subjects, 16 patients with angina pectoris, and 45 patients with hypoxic chronic bronchitis and emphysema. The mean RVEF in the control subjects was 0.62 + 0 09 (SD), which was not significantly different from the mean RVEF in the patients with angina (0.60 + 0.09), but was significantly higher (p < 0.01) than the mean value in patients with chronic bronchitis and emphysema (0.45 + 0. 11). LVEF was not significantly different in the groups studied. There was a significant correlation between LVEF and RVEF only in patients with chronic bronchitis and emphysema (p < 0.001). Those patients with chronic bronchitis and emphysema who had clinical evidence of cor pulmonale at the time of the study had significantly lower values of RVEF and LVEF (p < 0.001) than patients with no previous cor pulmonale or those who had had cor pulmonale in the past. There was a significant correlation between RVEF and arterial oxygen (p < 0.01) and carbon dioxide tensions (p < 0.05). Reduced RVEF in patients with chronic bronchitis and emphysema may be an early indicator of the development of cor pulmonale and may be useful as a non-invasive method of assessing the effects of therapeutic interventions. Right ventricular hypertrophy, with pulmonary hypertension, is a characteristic feature of the "blue and bloated" (type B, non-fighter) pattern seen in some patients with chronic bronchitis and emphysema.'-' Although the association between arterial hypoxaemia and pulmonary vasoconstric- tion is well known,5 the mechanism of development of cor pulmonale in these patients is unclear. Cor pulmonale is recognised clinically in patients with chronic bronchitis and emphysema by the presence of peripheral oedema for which no other cause can be found, and it is a poor prognostic sign, only one- third of such patients surviving for five years.4'6 The basis of the oedema in cor pulmonale remains in doubt.7 There is also debate about whether right heart function in such patients is normal or depre- Address for reprint requests: Dr W MacNee, Department of Medicine, Royal Infirmary, Edinburgh EH3 9YW. Accepted 7 March 1983 ssed.8 9 An alternative explanation of the mechanism of the oedema in clinical cor pulmonale suggests that there is a fall in renal blood flow that activates the renin-angiotensin-aldosterone system, resulting in sodium and water retention.'0-'3 The recent introduction of radionuclide methods of assessing right ventricular performance should improve our understanding of cardiac function in clinical cor pulmonale. Such measurements have previously been reported in patients with chronic bronchitis and emphysema.'415 Relatively few of these patients, however, have had severe distur- bances of arterial blood gas tensions or clinical cor pulmonale. In this study we have used multiple- gated equilibrium radionuclide ventriculography to measure right and left ventricular ejection fractions (RVEF, LVEF) in patients with severe chronic bronchitis and emphysema with and without cor pulmonale, and have related the findings to the clin- ical and electrocardiographic indices of cor pul- monale, lung function, and arterial blood gas ten- sions in these patients. 494 on October 2, 2021 by guest. Protected by copyright. http://thorax.bmj.com/ Thorax: first published as 10.1136/thx.38.7.494 on 1 July 1983. Downloaded from

Transcript of Assessment byradionuclide angiographyofright emphysema

Page 1: Assessment byradionuclide angiographyofright emphysema

Thorax 1983;38:494-500

Assessment by radionuclide angiography of right andleft ventricular function in chronic bronchitis andemphysemaW MACNEE, QF XUE, WJ HANNAN, DC FLENLEY, CJ ADIE, AL MUIR

From the Departments ofMedicine, Respiratory Medicine, and Medical Physics and Medical Engineering,University ofEdinburgh

ABSTRACT Non-invasive measurements of right and left ventricular ejection fraction (RVEF,LVEF) by multiple-gated equilibrium radionuclide ventriculography were performed in 18 con-trol subjects, 16 patients with angina pectoris, and 45 patients with hypoxic chronic bronchitisand emphysema. The mean RVEF in the control subjects was 0.62 + 0 09 (SD), which was notsignificantly different from the mean RVEF in the patients with angina (0.60 + 0.09), but wassignificantly higher (p < 0.01) than the mean value in patients with chronic bronchitis andemphysema (0.45 + 0.11). LVEF was not significantly different in the groups studied. There wasa significant correlation between LVEF and RVEF only in patients with chronic bronchitis andemphysema (p < 0.001). Those patients with chronic bronchitis and emphysema who had clinicalevidence of cor pulmonale at the time of the study had significantly lower values of RVEF andLVEF (p < 0.001) than patients with no previous cor pulmonale or those who had had corpulmonale in the past. There was a significant correlation between RVEF and arterial oxygen (p< 0.01) and carbon dioxide tensions (p < 0.05). Reduced RVEF in patients with chronicbronchitis and emphysema may be an early indicator of the development of cor pulmonale andmay be useful as a non-invasive method of assessing the effects of therapeutic interventions.

Right ventricular hypertrophy, with pulmonaryhypertension, is a characteristic feature of the "blueand bloated" (type B, non-fighter) pattern seen insome patients with chronic bronchitis andemphysema.'-' Although the association betweenarterial hypoxaemia and pulmonary vasoconstric-tion is well known,5 the mechanism of developmentof cor pulmonale in these patients is unclear. Corpulmonale is recognised clinically in patients withchronic bronchitis and emphysema by the presenceof peripheral oedema for which no other cause canbe found, and it is a poor prognostic sign, only one-third of such patients surviving for five years.4'6 Thebasis of the oedema in cor pulmonale remains indoubt.7 There is also debate about whether rightheart function in such patients is normal or depre-

Address for reprint requests: Dr W MacNee, Department ofMedicine, Royal Infirmary, Edinburgh EH3 9YW.

Accepted 7 March 1983

ssed.8 9 An alternative explanation of the mechanismof the oedema in clinical cor pulmonale suggests thatthere is a fall in renal blood flow that activates therenin-angiotensin-aldosterone system, resulting insodium and water retention.'0-'3The recent introduction of radionuclide methods

of assessing right ventricular performance shouldimprove our understanding of cardiac function inclinical cor pulmonale. Such measurements havepreviously been reported in patients with chronicbronchitis and emphysema.'415 Relatively few ofthese patients, however, have had severe distur-bances of arterial blood gas tensions or clinical corpulmonale. In this study we have used multiple-gated equilibrium radionuclide ventriculography tomeasure right and left ventricular ejection fractions(RVEF, LVEF) in patients with severe chronicbronchitis and emphysema with and without corpulmonale, and have related the findings to the clin-ical and electrocardiographic indices of cor pul-monale, lung function, and arterial blood gas ten-sions in these patients.

494

on October 2, 2021 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.38.7.494 on 1 July 1983. D

ownloaded from

Page 2: Assessment byradionuclide angiographyofright emphysema

Ventricular performance in chronic bronchitis and emphysema

Methods

SUBJECTSWe have measured right and left ventricular ejectionfractions by multiple-gated equilibrium blood poolscintigraphy in 18 control subjects (14M, 4F, aged21-63 years, mean 42 years), in 16 patients withangina (13M, 3F, aged 46-72 years, mean 58 years),and in 45 patients with chronic bronchitis andemphysema (29M and 16W, aged 40-81 years,mean 59 years). The patients with chronic bronchitisand emphysema also had RVEF measured by first-pass radionuclide angiography. All 18 control sub-jects had been referred for investigation of atypicalchest pain but had no history or clinical evidence oflung disease, and their chest radiographs and theirelectrocardiograms and radionuclide LVEF, both atrest and on-exercise, were al;l nrormal. None of thesecontrol subjects was receiving respiratory or car-diovascular drugs at the time of study, and the sub-sequent clinical course showed that none hadischaemic heart disease or chronic bronchitis. The45 patients with chronic bronchitis and emphysemahad irreversible airflow limitation (FEV, 0 7, SD +0-3 1 (range 0.4-2.0); forced ventilatory capacity(FVC) 2.1 08 1 (range 0.5-3-7)) and hypoxaemia(Pao2 6-9 + 1.3 kPa (52 10 mm Hg), range 4.7-9.6 kPa (35-72 mm Hg)), and most had a compen-sated respiratory acidosis (PaCo2 6-9 + 1*2 kPa (52+ 9 mm Hg), range 4*5-9-2 kPa (34-69 mm Hg);arterial H+ activity 42 + 4 nmol/l, range 35-51nmol/l). The 16 patients with angina pectoris had ahistory typical of angina and electrocardiographicevidence of ischaemic heart disease, but were not inheart failure and had not suffered previous myocar-dial infarction or lung disease. In all patients currentdrug treatment was continued during the study butbronchodilators were discontinued four hoursbefore measurement of FEV, and FVC. Arterialblood gas tensions, FEV, and FVC were measuredwhile they were breathing room air; and electrocar-diograms were done within 24 hours of theradionuclide study. In 12 patients pulmonary arter-ial pressure and cardiac output were measured at thesame time as the radionuclide study was carried out,a Swan-Ganz flow-directed balloon catheter beingused. Cor pulmonale was diagnosed clinically bydocumented evidence of ankle oedema, for whichno other cause could be found, that was present atthe time of study or had been found in the past. Onthis basis 12 patients had no cor pulmonale, 19 hadhad it in the past, and 14 had it at the time of study.Fourteen patients had received domiciliary oxygenfor between six months and five years, for 15 hours a

day at flow rates of 1-2 /min by nasal prongs.'6 Fourpatients had RVEF measured before and six months

after 15 hours a day of such oxygen treatment. Allmeasurements of RVEF and LVEF were madewhen the patients were breathing room air. Thestudy was approved by the hospital ethical commit-tee and the Medical Research Council's IsotopeCommittee, and each patient gave informed consentto the study.

RADIONUCLIDE TECHNIQUESThe details of the radionuclide techniques aredescribed elsewhere.'7 18 In brief, in all studies thesubjects lay supine beneath a Searle LEM gammacamera connected to a microcomputer (CromencoSystems 3). Heart rate was derived from the ECG,which was monitored continuously throughout thestudy, and used to trigger, or gate, the ventriculo-gram. Initially, for the first-pass measurement, thegamma camera was positioned in a 200 left anterioroblique projection with a 100 caudal tilt; thereafterthe position was changed to a 200 left anterior obli-que view with a 20° caudal tilt to optimise separationof the right and left ventricles and right atrium, assuggested by our previous studies.'8 Radioactivityfrom 500 sequential cardiac cycles was recorded inframe mode. The images were displayed in "movie"format and regions of interest around the right andleft ventricles were chosen. After suitable back-ground correction, as described in our previouspaper,'8 the ejection fraction (EF) was calculatedfrom the equation

EF = EDC - ESCEDC

where EDC and ESC are the counts of radioactivitywithin the region of interest at end diastole and endsystole.

Values are given as means and standard devia-tions. Statistical comparisons were made with thepaired t test and analysis of variance.

Results

The correlation between RVEF values obtained bythe first-pass and the equilibrium methods in thepatients with chronic bronchitis and emphysema wasgood (r = 0-91, p < 0-001). As the measurement ofLVEF was made by the equilibrium method, wehave chosen to give the measurements of RVEFmade by the same method; but the results of thestudy do not differ significantly whichever of the twomethods of obtaining the RVEF values is used.

RIGHT VENTRICULAR EJECTION FRACTIONThe mean RVEF in 18 control subjects was 0-62 ±

495

on October 2, 2021 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.38.7.494 on 1 July 1983. D

ownloaded from

Page 3: Assessment byradionuclide angiographyofright emphysema

-o RVEFm LVEF

a o

m A ,~~~

0

n=18 n=16 n=42 n=45

Control Angina CB and E

Fig 1 Right pnd left ventricular ejection fraction (RVEF,LVEF) in control subjects and patients with angina orchronic bronchitis and emphysema (CB and E). The pointsare individual values forRVEF (0) and LVEF (U) and themean values for each group (a) t I SD are shown. Themean RVEF in patients with chronic bronchitis andemphysema is significantly lower than in controls (p <0.01).

0*09 (range 0.50-0-83) (fig 1). The mean RVEF inthe 16 patients with angina was 0-60 + 0-09 (range0-50-0.80), which was not significantly differentfrom the value in control subjects. The mean RVEFin patients with chronic bronchitis and emphysema,

MacNee, Xue, Hannan, Flenley, Adie, Muir

however, was 0-45 + 0.11 (range 0-24-0.66), whichwas significantly lower (p < 0-01) than the mean forthe control subjects. Of the 45 patients with chronicbronchitis and emphysema, 29 had RVEF valuesbelow 0-50.

LEFT VENTRICULAR EJECTION FRACTIONThere was no significant difference between meanvalues of LVEF in control subjects (0-64 ± 0-08),patients with angina (0-60 ± 0-11), and patientswith chronic bronchitis and emphysema (058 +0-13) (fig 1). The lowest value of LVEF in our con-trol subjects was 0*51. Six patients with angina and16 patients with chronic bronchitis and emphysema(only one of the latter having electrocardiographicevidence of ischaemic heart disease) had an LVEFbelow 0-50.The range of values of RVEF and LVEF was

much wider in patients with chronic bronchitis andemphysema than in those with angina and the con-trol subjects. In neither the control subjects norpatients with angina did values of RVEF correlatewith LVEF (r = 0*275; p > 0.1), whereas RVEFwas significantly correlated with LVEF (r = 0-61; p< 0.01) in the patients with chronic bronchitis andemphysema (fig 2). As expected, those patients withchronic bronchitis and emphysema who either hadcor pulmonale at the time of study or had had it inthe past (as defined previously) had values of FEV,

,- 0 e

0 0

0*0 .

*1. .00

0*-

* -

,-. 0

0

0*

0..

0

0

00 - ,- ,

Fig 2 The correlationbetween right and leftventricular ejectionfraction (RVEFandLVEF) in 43 patientswith chronicbronchitis andemphysema (r =0-61,p <0.01). The solidline is the kast-squaresfit to the data and thebroken lines are the9S% confidence limits.

I I I I I I0.40 0.60

Right ventricular ejection fraction

496

1.0

0-80~C

0

'4-c0

4)a)'4

Cd4,

0 60_

0 401_

0 20

1.0r

0.80k

0.60p

C0

4-u

4)4).2_

-

c

14)4)-J

0o20F

U 0u20 n'r% 1%^

01

"I.1

1-1,l

.11,l

1111

I

u bu0

I

i 4 0

1m

on October 2, 2021 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.38.7.494 on 1 July 1983. D

ownloaded from

Page 4: Assessment byradionuclide angiographyofright emphysema

Ventricular performance in chronic bronchitis and emphysema

and Pao2 which were significantly lower (p < 0.01)and values of Paco2 significantly higher (p < 0.01)than those patients without cor pulmonale (table 1).RVEF was lower (p < 0.001), however, only inthose who had cor pulmonale at the time of study,

- 0 RVEF* LVEF

I

R8 i

8

n=12

NCP

Fig 3 Right and left ventricular ejand LVEF) in patients with chroniemphysema who had no cor pulmthad cor pulmonale in the past (PCpulmonale at the time ofstudy (C]individual values for RVEF (0) aivalues for each group (0) 1I SDNS; NCPv CP,p <0.01).

(a)

l '__~~~[ 0. _

I~~~~~

li.i r |: 40~~

Table 1 Blood gas tensions and FEV, in 45 patients withchronic bronchitis and emphysema

Padent group PaO2 PaCO2 FEV,(kPa) (kPa) (1)

Nocorpulmonale 8-1 + 1-3 6-2 + 0-8 0-9 + 0-4Previous corpulmonale 6-7 + 1.2*** 6-9 - 1.3** 0-7 - 0-3**

Cor pulmonalepresent 6-2 + 0-9*** 7-3 - 1-0** 0-6 0-3**

For the definition of cor pulmonale see text. The data are means +1 standard deviation. * = p < 0-05; ** = p < 0-01; *** = p <0.001. The p values relate to significant differences from the resultsin those with no cor pulmonale.

Conversion: SI to traditional unit-Arterial oxygen tension (Pao2)and arterial carbon dioxide tension: 1 kPa = 7-5 mm Hg.

n=19 n=12 n=14 and not in those patients who had had it in the past.None of the 14 patients with cor pulmonale at the

PCP cP time of study had a normal RVEF (> 0-50) whereasiccbronchiractiond(RVE nine of the 19 patients with past cor pulmonale hadictironcfrationd(VE a low value of RVEF (fig 3).onale (NCP), who had LVEF was significantly lower (p < 0-001) in those'P), and who had cor patients with cor pulmonale at the time of study thanP). The points are in those with no previous cor pulmonale and thosend LVEF (U). The mean who had had it previously (fig 3). In 16 patients withareshown (NCP v PCP, chronic bronchitis and emphysema LVEF was lower

0

~.

*: 0

* 00

0.00

n= 31 --.

r =0-35 -

p<0.05

(b)10 _

8 _

6a-

00~

4

2

I nu

0.20 0.40 0.60 0 80 10 0.20Right ventricular ejection fraction

I/

. A 0

A *,A-.

-

.,-A~~~~~~~0.1* 0 *, A0 A

01

.01-

/

n=31r =051p <0-01

I

0.40 0.60 0.80 10

Fig 4 Correlation between right ventricular ejection fraction (RVEF) and (a) arterial oxygen tensions (Pao) and (b)arterial carbon dioxide tensions (Paco) measured when they were breathing air in 31 patients with chronic bronchitis andemphysema. The solid lines are the least-squares fit to the data and the broken lines are the 95% confidence limits.

c

10

0

a4,

0.801_

0.601_

0 401-

0.201_

10

8

6

2

uL.._

u.- U. I

497

'-r

--_

on October 2, 2021 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.38.7.494 on 1 July 1983. D

ownloaded from

Page 5: Assessment byradionuclide angiographyofright emphysema

MacNee, Xue, Hannan, Flenley, Adie, Muir

0.80

0

U6C

0

4),

0::

0.60kS

0

Uso

0.40k

0.20kn=20

{0

ITIISOLSI

*

0

n= 23

0

S

SI

0

n= 20

Fig 5 Comparison oftheelectrocardiographicfindings and right

a ventricular ejectionfraction (RVEF) in 45

T patients with chronicbronchitis and emphysema.The points are individualvalues (0), with the mean

a RVEFfor eachelectrocardiographic

* category (0) ± 1 SD.

n=9

nNormal Right axisECG deviation

'P'pulmonale

than normal (< 0-50). None of these had hyperten-sion; one had ischaemic heart disease, eight had corpulmonale at the time of study, six had had it in thepast, and only two had had no previous cor pul-monale. All patients with a low LVEF also had a lowRVEF.

In the patients with chronic bronchitis andemphysema there was no significant correlationbetween RVEF and either FEV, (r = 0.16), percen-tage of predicted FEV, (r = 0.17), or FVC (r =0 23) (p > 0-1 in each case). In 12 patients in whompulmonary arterial pressure was measured at thetime of the radionuclide study RVEF did not corre-late with the mean or the systolic pulmonary arterialpressure (r = 0 15, p > 0-5) or the pulmonary vascu-lar resistance (r = 0-09, p > 0-5). There was, how-ever, a significant correlation between RVEF andPao, (r = 0-51, p < 0-01) and between RVEF andPaco2 (r = -0X35, p < 0X05), but not betweenRVEF and arterial hydrogen ion concentration (r =0-15, p > 0X05) (fig 4) in 31 patients who were notreceiving long-term oxygen treatment (RVEF 0X44+ 10, Pao2 7-3 + 1-3 kPa (54.6 + 9-6 mm Hg)).There was no significant correlation between RVEFand Paco2 and Paco2 in the 14 patients who werereceiving long-term oxygen treatment-(RVEF 0-51+0-11, Pao2 6-2 + 1-0 kPa (46-5 + 7-5 mm Hg)) (r=0o09, p> 0.1).In the patients with chronic bronchitis and

emphysema there was a significant correlation be-tween LVEF and Pao, (r = 0.64, p < 0-001) in the31 patients who were not receiving long-termoxygen treatment, but no correlation betweenLVEF and any of the other variables.

Right ventricularhypertrophy

Four patients were studied before and six monthsafter the start of treatment with low-flow oxygen for15 hours in every 24 hours. Three patients who hadan abnormal RVEF (mean 0-35) before oxygentreatment had a normal RVEF (mean 0.53) after sixmonths' treatment. One patient who had a normalRVEF (0.55) showed no change in RVEF after thisperiod of oxygen treatment.

Electrocardiographic evidence of cor pulmonale(fig 5) did not distinguish patients with normal andlow values of RVEF.

Discussion

The complex geometry of the right ventricle makesangiographic assessment of right ventricular vol-umes difficult, and although stroke volumes deter-mined by the angiographic and the thermodilutiontechniques have been shown to be correlated'9 con-trast angiography has not been widely applied in theassessment of right ventri6ular performance. Similargeometric constraints apply to attempts to analyseright ventricular performance by echocardiography.As radionuclide techniques do not depend on aknowledge of the geometry of the cardiac chambers,and volumes of the chambers are directly propor-tional to radioactive counts, these methods havegreat attraction in the assessment of right ventricularperformance. Most radionuclide studies of the rightventricle have been carried out with a first-passtechnique, which allows optimal separation betweenright atrium and right ventricle. Count rate limita-tions, however, make this technique less suited toconventional gamma cameras. We have modified

U'

498

on October 2, 2021 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.38.7.494 on 1 July 1983. D

ownloaded from

Page 6: Assessment byradionuclide angiographyofright emphysema

Ventricular performance in chronic bronchitis and emphysema

the equilibrium blood-pool technique first describedby Maddahi,'7 and have shown that results correlatewell with those obtained with the first-passmethod.'8 In the presentatiion of results in this studywe have used the measurements derived from theequilibrium technique, as the left ventricular ejec-tion fraction was also determined by this technique.The results would be virtually identical, however, ifthe values obtained from the first-pass techniquehad been used. The range of right ventricular ejec-tion fractions in our 18 control subjects was from0 50 to 0-83, similar to values in previous controlsubjects obtained by a first-pass technique,'4 15 20 buthigher than the normal range obtained by Maddahi'7(0.390-57) using an equilibrium technique. OurLVEF values were similar to the results obtained incontrol subjects by either first-pass or equilibriummethods.415 172021 We found no significant differ-ence in either right or left ventricular ejection frac-tion between control subjects and those with anginapectoris.We can confirm that right ventricular ejection

fraction varies widely in chronic bronchitis andemphysema, but is significantly lower than in controlsubjects, thus confirming the results of earlierstudies.'4 15 20 All of our patients with chronic bron-chitis and emphysema were hypoxic (Pao2 6-9 + 1*3kPa (51.6 + 9-8 mm Hg)), and 33 of the 45 patientswere hypercapnic. Most (33) had cor pulmonale atthe time of study or had had it previously. Despitehaving these severe abnormalities of respiratoryfunction 16 of the patients had a normal right ven-tricular ejection fraction, suggesting that right ven-tricular function (as reflected by this measurement)is maintained until a late stage of the disease. Theleft ventricular ejection fraction was normal in mostof the patients with chronic bronchitis andemphysema but was below normal in 16 of the 45patients. Previous workers have not reported a cor-relation between the ejection fractions of the rightand left ventricles in patients with chronic bronchitisand emphysema,'42223 but this may be related toselection of patients. Berger et al studied patientswith chronic obstructive lung disease, of whom 28%had cor pulmonale'4; whereas in this present study73% had cor pulmonale. This, coupled with the factthat abnormalities of arterial blood gas tensions inour patients were more severe, could explain whythe correlations between right and left ventricularejection fraction were more obvious. Differences indefinition of the clinical syndrome of cor pulmonaleare important in this context. Whereas Berger madeECG evidence of right ventricular hypertrophy aprerequisite for this diagnosis,14 we have notassumed that cor pulmonale exists only with thiselectrocardiographic feature in patients with chronic

bronchitis and emphysema. Pitting ankle oedema,for which no other cause can be found, in a patientwith chronic bronchitis and emphysema is the basisof our clinical diagnosis of cor pulmonale as the levelof the jugular venous pressure is less reliable inpatients with severe airflow obstruction. By thiscriterion we found that the electrocardiographicevidence of right ventricular hypertrophy did notseparate those patients with clinical cor pulmonalefrom those without cor pulmonale, nor did it corre-late with the right ventricular ejection fraction. Thelack of correlation between electrocardiographicfeatures and clinical evidence of cor pulmonale hasbeen noted by previous workers.324

Previous studies with invasive techniques havepresented conflicting data on left ventricular func-tion in chronic bronchitis and emphysema.825 Ourdata suggest that the degree of depression of bothright and left ventricular function relates to arterialhypoxaemia, but this applies only to patients whowere not receiving long-term oxygen treatment(arterial blood gas tensions and all radionuclidestudies were performed when the patients werebreathing room air). The lack of correlation bet-ween right ventricular ejection fraction and Pao2and Paco2 in patients receiving long-term oxygentreatment may be due to the higher Pao2 attained bythese patients for 15 hours of each day. Althoughprevious workers'4 's have not distinguished patientswho were receiving long-term oxygen, this groupmay be a separate population as the treatment isknown to affect pulmonary haemodynamics.'6 26 Ourresults suggest that Paco2 and to a lesser extent (inthe case of right ventricular ejection fraction) Paco2are more important in determining right and leftventricular ejection fraction in patients with chronicbronchitis and emphysema. Our preliminary resultson the effects of long-term oxygen on right ventricu-lar ejection fraction in this study are similar to thosereported by Ellis et al27 and are encouraging,although there is disagreement about whether theimprovement produced by oxygen treatment is aresult of acute reduction of right ventricular after-load'627 or due to a more long-term effect of suchtreatment on pulmonary arterial structure.

Ejection phase indices are dependent on both thepreload and afterload on the ventricle and pulmo-nary hypertension would be expected to reduce theright ventricular ejection fraction. We were sur-prised that we could not correlate changes in rightventricular ejection fraction with pulmonary arterialpressure or pulmonary vascular resistance in 12 ofour patients with chronic bronchitis andemphysema. Similar findings, however, have beenreported by Ellis et al.27 An increase in pulmonaryarterial pressure could be the main contributor to

499

on October 2, 2021 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.38.7.494 on 1 July 1983. D

ownloaded from

Page 7: Assessment byradionuclide angiographyofright emphysema

500

the reduction in ejection fraction in acute cor pul-monale, but the left ventricle is not then subjected toan increased afterload and this would not explainthe reduction in left ventricular ejection fraction inour patients with cor pulmonale at the time of study.

In conclusion, we have used gated equilibriumblood-pool radionuclide ventriculography to showthat the right ventricular ejection fraction is reducedin patients with chronic bronchitis and emphysemaand is further reduced, along with a concomitantreduction in left ventricular ejection fraction, inthose patients who have cor pulmonale at the timeof study. As cardiac output and stroke volume canalso be measured by radionuclide methods28 abso-lute determinations of end-diastolic and end-systolicvolumes may be made, and by relating these tointracardiac pressures the mechanics of the rightventricle in cor pulmonale can be understood more

clearly.

References

'Dornhorst AC. Respiratory insufficiency. Lancet1955;i:1185-7.

2Thomas AJ. Chronic pulmonary heart disease. Br HeartJ 1972;34:653-7.

3Fishman AP. Chronic cor pulmonale.Am Rev Respir Dis1976;114:775-94.

4 Renzetti AD jun, McClement JH, Litt BD. The Vete-rans Administration Co-operative Study on Pulmo-nary Function. III Mortality in relation to respiratoryfunction in chronic obstructive pulmonary disease.Am J Med 1966;41:115-29.

5Weitzenblum E, Loiseau A, Hurth C, Mirhom R,Rasaholinjanahory J. Course of pulmonaryhaemodynamics in patients with chronic obstructivepulmonary disease. Chest 1979;75:656-62.

6 Middleton HC, Peake MD, Howard P. Hypoxaemia inchronic bronchitis. Thorax 1979;34:313-6.

7Richens JM, Howard P. Oedema in cor pulmonale. CliSci 1982;62:255-9.

8 Khaja F, Parker JO. Right and left ventricular perfor-mance in chronic obstructive lung disease. Am Heart J1971 ;82:319-27.

Jezek V, Schrijen F. Left ventricular function in chronicobstructive lung disease with and without cardiac fail-ure. Cli Sci Mol Med 1973;45:267-79.

'0Platts MM, Hammond JDS, Stuart-Harris CH. A studyof cor pulmonale in patients with chronic bronchitis. QJ Med 1960;29:559-74.

Aber GM, Bishop JM. Small changes in renal function,arterial gas tensions and acid-base state in patientswith chronic bronchitis and oedema. Cli Sci1965;281:511-25.

12Tomaszewski J, Kowalewski J, Wrukowska M. Plasmaactivity in patients with chronic cor pulmonale syn-drome. Polish Medical Science and History Bulletin

MacNee, Xue, Hannan, Flenley, Adie, Muir

1975;15:207-11.3 Stuart-Harris CH, Mackinnon J, Hammond JDS, Smith

WD. The renal circulation in chronic pulmonary dis-ease and pulmonary heart failure. Q J Med1956;25:389405.

'4Berger HJ, Matthay RA, Lohe J, Marshall RC,Gottschalk A, Zaret BL. Assessment of cardiac per-formance with quantitative radionuclide angiography.Right ventricular ejection fraction with reference tofindings in chronic obstructive pulmonary disease.AmJ Cardiol 1978;41:897-905.

Slutsky RA, Acherman W, Karliner JS, Ashburn WL,Moser KM. Right and left ventricular dysfunction inpatients with chronic obstructive lung disease.Assessment by first-pass radionuclide angiography.Am J Med 1980;68:197-205.

16Medical Research Council. Long-term domiciliaryoxygen therapy in chronic hypoxic cor pulmonalecomplicating chronic bronchitis and emphysema. Lan-cet 1981;i:681-6.

7 Maddahi J, Berman DS, Dale T, et al. A new techniquefor assessing right ventricular ejection fraction usingrapid multiple-gated equilibrium cardiac blood poolscintigraphy. Circulation 1979;60:581-9.

18Xue QF, MacNee W, Flenley DC, Hannan WJ, Adie CJ,Muir AL. Can right ventricular performance be asses-sed by equilibrium radionuclide ventriculographyThorax; 38:486-93.

4 Gentzler RD, Briselli MF, Gault GH. Angiographicestimation of right ventricular volume in man. Circula-tion 1974;50:324-30.

20 Matthay RA, Berger HJ, Davies RA, et al. Right and leftventricular exercise performance in chronic obstruc-tive pulmonary disease. Radionuclide assessment.Ann Intern Med 1980;93:234-9.

21 Burrow R, Strauss HW, Singleton R, et al. Analysis ofleft ventricular function from multiple-gated acquisi-tion cardiac blood pool imaging: comparison to con-trast angiography. Circulation 1977;56:1024-8.

22 Steele P, Ellis J, Van Dyke D, Sutton F, Creagh E, DavisH. Left ventricular ejection fraction in severe obstruc-tive lung disease. Am J Med 1975;59:21-7.

23 Kline L, Crawford M, MacDonald W, Schelbert H,O'Rourke R. Non-invasive assessment of left ven-tricular performance in patients with chronic obstruc-tive lung disease. Chest 1977;72:558-64.

24 Padmavati S, Raizada V. The electrocardiogram inchronic cor pulmonale. Br Heart J 1973;34:658-67.

25 Burrows B, Kettel L, Nides A, et al. Patterns of car-diovascular dysfunction in chronic obstructive lungdisease. N Engl J Med 1972;286:912-7.

26 Nocturnal Oxygen Therapy Trial Group. Continuous ornocturnal oxygen therapy in hypoxaemic chronic obs-tructive lung disease.Ann Intern Med 1980;93:391-8.

27 Ellis JH Jr, Kirch D, Steele PP. Right ventricular ejec-tion fraction in severe chronic airway obstruction.Chest 1977;71:281-2.

28Hannan WJ, Vojacek J, Dewhurst NG, Muir AL. Thesequential measurement of ventricular volumes andcardiac output by radionuclides. Clinical, Physical andPhysiological Measurement 1980;1:125-34.

on October 2, 2021 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.38.7.494 on 1 July 1983. D

ownloaded from