Clinical use of ultrashort-lived radionuclide krypton-81m for noninvasive analysis of right ventricular performance in normal subjects and patients with right ventricular dysfunction

687

JACC Vol. 5, No.3 March 1985:687-98

METHODS

Clinical Use of Ultrashort-Lived Radionuclide Krypton-81m for Noninvasive Analysis of Right Ventricular Performance in Normal Subjects and Patients With Right Ventricular Dysfunction CHRISTOPH A. NIENABER, MD, ROLF P. SPIELMANN, MD,* GERD WASMUS, PHD,* DETLEF G. MATHEY, MD, RICARDO MONTZ, MD,* WALTER H. BLEIFELD, MD, FACC Hamburg, West Germany

The ultrashort-lived radionuclide krypton-81m, eluted in 5% dextrose from a bedside rubidium-81m generator, was intravenously infused for rapid imaging of the rightsided heart chambers in the right anterior oblique projection adjusted for optimal right atrioventricular separation. Left-sided heart and lung background was minimized by rapid decay and efficient exhalation of krypton. 81m, requiring no algorithm for background correction. A double region of interest method decreased the variability in the assessment of ejection fraction to 5 % . In 10 normal subjects, 11 patients with pulmonary hypertension, 4 patients with right ventricular outflow tract obstruction and 4 patients with right ventricular infarction, right ventricular ejection fraction determined by krypton-81m equilibrium blood pool imaging ranged from 14 to 76%. The correlation between these values and those determined by cineangiography according to Simpson's rule was close: r = 0.93 for all data points

In 1968, the ultrashort-lived radionuclide krypton-81m was introduced as an imaging agent by Yano and Anger (1). Krypton-81m minigenerators were designed (2) to provide elution of this noble gas from rubidium-81 m for the assessment of lung ventilation and perfusion (3,4), cerebral blood flow (3,5) and regional myocardial perfusion in animals and human patients (6-8). Recently, krypton-81m was used to assess right ventricular function in human beings (9). This study describes the clinical application of krypton81m to image the right ventricular blood pool and determine right ventricular ejection fraction, both in normal subjects and patients with right ventricular dysfunction at rest and

From the Department of Cardiology and Department of Nuclear Medicine,* University Hospital Eppendorf, Hamburg. West Germany. Manuscript received February 29, 1984; revised manuscript received September 24, 1984, accepted October 15, 1984. Address for reprints: Christoph A. Nienaber, MD, University Hospital Eppendorf, Department of Cardiology, 2. Med. Clinic, Martinistr. 52. 2000 Hamburg 20, West Germany. © 1985 by the American College of Cardiology

(p < 0.001), r = 0.92 for studies at rest (p < 0.001) and r = 0.93 for exercise studies (p < 0.(01). Exerciserelated changes in right ventricular function revealed a disturbed functional reserve with pulmonary hypertension and right ventricular infarction, whereas in compensated right ventricular outflowtract obstruction there was a physiologic increase in ejection fraction with exercise (p < 0.001). Thus, equilibrium-gated right ventricular imaging using ultrashort-lived krypton-81m is a simple, accurate and reproducible method with potential for serial assessment of right ventricular ejection fraction in a variety of right ventricular anatomic and functional abnormalities, both at rest and during exercise. Advantages of this method include an extremely low radiation dose to patients and clear right atrioventricular separation without the need to correct for background activity. (J Am Coli CardioI1985;5.·687-98)

during exercise. The method was validated by cineangiography and tested for reproducibility. Other noninvasive methods have been proposed to measure right ventricular ejection fraction using technetium-99m (half-life of 5.6 hours) in first pass acquisition mode independent of right ventricular geometry. These techniques compare well with biplane cineangiography (10-19). However, since only the first transit of technetium-99m is recorded, the technique does not provide serial measurements without repeated injections of technetium-99m for each data acquisition point. The number of repeat first pass studies, however, is limited by increasing radiation exposure to the patient and by residual blood pool background from previous injections. To eliminate problems associated with first pass technetium-99m angiography, multiple gated equilibrium blood pool imaging in left anterior oblique projection had been introduced and validated to measure right ventricular ejection fraction assuming counts to be proportional to volume (18-21). Problems remain , however, in defining right 0735-1097/85/$3.30

688

NIENABER ET AL. RIGHT VENTRICULAR IMAGING WITH KRYPTON-81m

ventricular background and managing right atrial overlap. The proposed methods thus differ largely in their approach to solving these problems (20,22). In this study, the ultrashort-lived radionuclide krypton81m is continuously infused intravenously to image exclusively the right-sided cardiac blood pool since rapid decay and low water solubility of krypton-81 m result in complete clearance of activity from the alveolar membrane. Thus, the technique provides serial measurements of right ventricular ejection fraction in a gated equilibrium acquisition mode using the right anterior oblique projection for optimal separation of the right atrium from the right ventricle, with no need to correct for lung background and minimal radiation burden to the patients.

Methods Krypton-81m generator. The generator consists of rubidium-81, produced in a cyclotron by alpha bombardment of bromine-79. Rubidium-81 decays with a half-life of 4.7 hours to ultrashort-lived krypton-81m, which has a half-life of 13 seconds; this process emits monoenergetic photons with an energy of 190 keV suitable for standard scintillation cameras and collimators (23,24). Krypton-81m is eluted by passing a solution of nonionic 5% dextrose in water through a cation exchange column in which the parent nuclide rubidium-81 is bound (25-27). The aqueous krypton-81 m solution is continuously recovered from the minigenerator and intravenously infused by a high precision pump (LKB HPLC 2150) providing a steady flow of 4 mllmin through a millipore filter system (Miller, Millipore SA) connected to an 18 gauge polyethylene cannula. Sterility of the eluate was proven by negative broth cultures. Rubidium-81 m breakthrough measured with gamma ray spectroscopy using a solid state detector system is insignificant (25). The charged generator is shielded by 9.5 em lead bricks completely surrounding the system. The activity level in the shielding surface is several milliroentgens/hour, causing no interference with imaging krypton-81 m when positioned 20 cm lateral to the collimator. This krypton81m minigenerator is not yet commercially available. Provided by the Deutsche Kemforschungszentrum Karlsruhe, its use is limited to about 18 hours, including transport time, because of the 4.7 hour half-life of the 25 mCi rubidium81m parent nuclide. Study patients (Table 1). Both krypton-81 m radionuclide and cineangiographic right ventricular ejection fraction studies were performed in 28 patients. Group 1consisted of 10 normal subjects, 7 men and 3 women, aged 25 to 56 years (mean 38), with no clinical or hemodynamic evidence of cardiopulmonary disease by physical and electrocardiographic examination, chest X-ray film and right heart catheterization. These 10 subjects were considered controls since

JACC Vol. 5, No.3 March 1985:687-98

results of both right heart catheterization and exercise radionuclide imaging studies for diagnostic evaluation of suspected cardiac disease were normal. Group II consisted of 18 patients (13 men and 5 women), aged 24 to 66 years (mean 45), with predominantly right ventricular abnormalities, and included 9 patients with secondary pulmonary hypertension (6 of whom had mild tricuspid valve regurgitation). One patient had primary pulmonary hypertension, three patients had partial correction of tetralogy of Fallot and residual pulmonary valve stenosis, one patient had isolated subvalvular pulmonary stenosis and four patients had complete occlusion of the right coronary artery and inferior myocardial infarction partly involving the right ventricle. Eight of the nine patients with secondary pulmonary hypertension had previous mitral valve replacement; four of these underwent both aortic and mitral valve replacement. Three patients in this subset had evidence of chronic obstructive pulmonary disease. Written informed consent according to the Helsinki declaration on human experimentation was obtained from all subjects undergoing either the angiographic or the radionuclide studies. Both studies were completed within 72 hours without changes in ongoing medication. There were no severe complications associated with either diagnostic study in any subject with the exception of Patient 4, who suffered from exhaustion and dizziness after stress cineangiography; Patients 9 and 10 could not perform dynamic exercise because of dyspnea in the supine position. Radionuclide angiography. Equilibrium radionuclide right ventricular angiography at rest and during submaximal exercise was performed with the patient in the supine position using a variable right anterior oblique view adjusted by 5 to 10° caudal tilt for best separation between the right atrium and the right ventricle. A mobile single crystal gamma scintillation camera (LEM, Searle) equipped with a low energy, all purpose, parallel-hole collimator was used for image acquisition, with the pulse-height analyzer set at 190 keY and a 20% acoustical window. Multiple gated acquisition was controlled by a PDP 11/34 minicomputer using GAMMA-II software. Thirty frames per cycle were acquired in frame mode by use of a zoom device that expanded the data 1.7 times and recorded it on 64 x 64 matrix. Acquisition at rest was terminated when 3,000,000 counts were reached in the total field of view (10 minutes). Exercise acquisition time was limited to a maximal period of 300 seconds, yielding 1,500,000 counts at a count rate of 5,000 counts/s ± 10%. Approximately 500,000 counts were routinely obtained in the end-diastolic region of interest. Data were processed using a semiautomated computer algorithm that involved definition of separate end-systolic and enddiastolic right ventricular regions of interest. Nine point spatial and three point time-smoothing was performed on each frame of a composite cycle. Operator intervention was permitted to define the tricus-

689

NIENABER ET AL. RIGHT VENTRICULAR IMAGING WITH KRYPTON-81m

JACC Vol. 5, No.3 March 1985:687-98

Table 1. Mean and Individual Hemodynamic Results in 10 Normal Subjects and 18 Patients With Right Ventricular Abnormalities Group I: 10 Normal Subjects OM, 3F; mean age 38 years)

CO (liters/min)

Ex

Re

Ex

Re

4 6 ±3 ±3 P < 0.05

5 ±3

8 ±4 NS

RAP (mm Hg)

Re

Re

Ex

13 19 ±4 ±5 p < 0.05

Mean value ± SO p

RAV wave (mmHg)

PAP (mm Hg)

Ex

Simpson's RVEF

Kr-8lm RVEF

(%)

(%)

Re

Ex

Re

Ex

6.0 I\.4 ± \.0 ± \.9 P < 0.01

49 60 ±7 ±7 P < 0.001

52 63 ±8 ±7 P < 0.001

CO

Simpson's RVEF

Kr-8lm RVEF

(liters/min)

(%)

(%)

Group II: 18 Patients With Right Ventricular Abnormalities

Case Age (yr) (no.) & Sex

Diagnosis

I 30/M 2 29/M 3 61/M 4 6O/M 5 64/F 6 52/F 7 42/M 8 53/M 9 61/F 10 311M Mean value ± SO P (Re vs. Ex)

AVR, MVR, PH, TR MVR, PH MVR, capo, TR, AVR MVR, capo, PH, TR MVR, PH, TR MVR, PH, TR, AVR AVR, MVR, PH, TR PH,PE MVR, capo PPH

37/M II 12 24/M 13 24/M 14 31/F Mean Value ± SO p (Re vs. Ex) 15 45/M 16 60/M 17 54/F 18 66/M Mean value ± SO p (Re vs. Ex)

TOF, PS TOF, PS, PI TOF, PS PS

INF INF INF INF

+ + + +

aP:Re Ex 36 37 55 80 92 89 40 44 56 62 ±25 ±26 NS RVMI RVMI RVMI RVMI

NYHA Class

III III III lll/IV IV II

III111 II IV IV

0 0 0 0

0 0 III II

PCWP (mm Hg)

PAP (mm Hg)

RAP (mm Hg)

Re

Re

Re

Ex

Ex

22 25 52 55 40 48 14 19 II 34 22 36 22 30 39 49 17 19 45 55 20 32 53 18 15 50 12 16 22 15 42 6 18 26 76 15 49§ 36* 16* 23:j: ±5 ±6 ±19 ±7 p < 0.05 P < 0.01 7 8

8 II

7 7 ±I

10

10

±2 NS 10 13 12 12 15 18 14 16 13* 15* ±2 ±3 NS

II 20 15 20 19 22 18 26 16 22 ±4 ±3 P < 0.05 18 22 13 18 16 29 14 30 IS 26 ±2 ±6 P < 0.05

Ex

RAV Wave (mm Hg) Re

Ex

Re

Ex

Re

Ex

Re

Ex

IS 16 II 13 7 15 16 20 15 19 10 21 9 20 I 8 9 13 II * 17t ±5 ±4 P < 0.01

20 23 3.2 4.9 12 16 3.4 5.4 15 24 4.6 9.4 19 34 3.9 4.9 18 26 5.4 6.9 13 29 12 38 6.5 8.1 3 17 4.4 6.4 IS 4.8 16 2.1 6.5§ 14t 26:j: 4.4* ±5 ±8 ±1.3 ± \.6 P < 0.01 P < 0.01

32 28 37 30 35 34 37 38 43 47 45 50 42 29 44 33 44 32 45 48 41 42 37 52 42 36 40 32 45 50 41 52 34 43 12 14 4!:j: 37* 37+ 40 ±IO ±8 ± 10 ±9 NS NS

5 12 9 10 8 II 10 13 8* 12t ±2 ±2 P < 0.05 6 9 10 12 12 14 10 II lOt 12t ±3 ±2 P < 0.02

6 12 13 II 10 ±3

43 56 45 62 50 56 53 69 50 65 51 69 42 49 46 54 46 56 49 64 ±4 ±7 ±4 ±7 P < 0.02 P < 0.01 40 51 43 48 41 36 36 39 29 33 30 28 36 40 38 38 35:j: 40:j: 38t 39§ ±5 ±8 ±6 ±9 NS NS

16 4.1 14 5.3 15 4.8 IS 4.7 15:j: 4.7 ±I ±0.5 NS P< 9 II 5.7 13 14 4.5 4.9 13 16 12 13 5.0 12t 14t 5.0 ±2 ±2 ±0.5 P < 0.05 P<

7.4 12.9 9.9 8.3 9.6 ±2.4 0.02 9.2 8.8 6.8 8.6 8.5:j: ± \.0 0.01

*p < 0.05; tp < 0.02; :j:p < 0.01; §p < 0.001 versus normal subjects. AVR = aortic valve replacement; CAD = coronary artery disease; CO = cardiac output; capo = chronic obstructive pulmonary disease; Ex = exercise; F = female; INF = inferior; M = male; MVR = mitral valve replacement; NS = not significant; ap = systolic pressure gradient of pulmonary stenosis; PAP = mean pulmonary artery pressure; PCWP = pulmonary capillary wedge pressure; PE = pulmonary embolism; PH = pulmonary hypertension; PI = pulmonary insufficiency; PPH = primary pulmonary hypertension; PS = pulmonary stenosis; RAP = mean right atrial pressure; Re = rest; RVEF = right ventricular ejection fraction; RVMI = right ventricular myocardial infarction; RVOT = right ventricular outflow tract; SO = standard deviation; TOF = tetralogy of Fallot; TR = tricuspid regurgitation.

pid and the pulmonary valve plane. If the right atrioventricular border could not be defined visually from original images, it was determined from phase images by first temperol Fourier transformation of time-activity curves (28,29) to compensate for systolic tricuspid valve plane motion. The pulmonary valve plane was defined as the junction between

the contracting and the noncontracting regions of the right ventricular outflow tract or the region of the outflow tract between a systolic increase in counts superiorly (pulmonary artery) and a systolic decrease in counts inferiorly. Right ventricular septal and free wall contours were aligned by a semi automated edge detection program; right ventric-

690

NIENABER ET AL. RIGHT VENTRICULAR IMAGING WITH KRYPTON-81m

ular end-diastolic and end-systolic regions of interest were defined by isocountlines of 50% of the right ventricular count maximum in the end-diastolic image. Right ventricular ejection fraction was calculated by subtracting endsystolic from end-diastolic counts divided by end-diastolic counts x 100. No background correction was performed. To test both methods (krypton-81 m ventriculography and contrast cineangiography) for inter- and intraobserver variability, two experienced observers (R.P.S. and C.A.N.), both blinded to the clinical diagnosis, independently evaluated ejection fraction in all subjects. In addition, for the assessment of intraobserver variability of each method, one observer blinded to his previous reading repeated the analysis I week after his first evaluation. Cineangiography (Table 2). Both rest and exercise biplane contrast ventriculography was performed in 16 of the 18 patients of Group II (Patients 9 and 10 were studied only at rest) and all 10 subjects in Group I in the anteroposterior and lateral views with power injection of 45 ml of meglumine diatrizoate (Urografin-76) at a rate of 15 mils and a pressure of 400 lb/in? (60 kg/em"). Stress right ventricular cineangiograms were performed at a level of submaximal exercise that was identical to the work load during the radionuclide study. Cine frames were obtained at a rate of 50 frames/so A I cnf grid was used for magnification correction. Maximal outward displacement of the right ventricle was determined in the end-diastolic frame, whereas the endsystolic frame was taken as the point of maximal inward motion. Simpson's rule was applied in the evaluation of both frames as described previously (12, 13) and used to calculate ejection fraction by subtracting the end-systolic volume from the end-diastolic volume divided by the enddiastolic volume x 100. Evaluation of hemodynamics. Systemic arterial blood pressure and heart rate were monitored conventionally at rest and during submaximal bicycle exercise. All patients who underwent cineangiography were studied simultaneously with invasive hemodynamic monitoring. Intracardiac pressures including pulmonary wedge pressure, mean pulmonary artery pressure and phasic right ventricular as well as phasic and mean right atrial pressures were obtained by right heart catheterization with a Swan-Ganz thermodilution catheter inserted through an antecubital or femoral vein using the Seldinger technique and advanced into the pulmonary artery. Cardiac output was determined by the thermodilution method. The pulmonary artery wedge pressure was measured as an index of left ventricular filling pressure. In four patients, the systolic pressure gradient between the right ventricle and pulmonary artery was used as an index for residual pulmonary stenosis. Exercise protocol. In each subject who underwent exercise studies, data were acquired at an identical work load. The patients pedaled at a constant speed, usually at a load

lACC Vol. 5, No, 3 March 1985:687-98

of 300 kpmlmin. Radionuclide imaging was started after 2 to 3 minutes or when a stable heart rate during exercise was reached. Hemodynamic measurements and cineangiography were also undertaken at this time. Patients 4 and 5 exercised to a work load of only 150 and 60 kpmlmin, respectively, for a period of 6 minutes; Patients 9 and 10 had no exercise studies (Table 2). Statistical methods. Ejection fraction values are reported as mean values ± standard deviation. Linear regression analysis was applied for statistical correlation between right ventricular ejection fraction by both methods. Student's t test was used to assess statistically significant differences. Rest and exercise data were compared by use of the paired t test. Intra- and interobserver variability of radionuclide and cineangiographic determinations of right ventricular ejection fraction is given as the residual error of a two-way analysis of variance. A probability (p) level of less than 0.05 was considered statistically significant.

Results Correlation between krypton-81m and angiographic right ventricular ejection fraction measurements. The correlation between the gated equilibrium technique using krypton-81m and biplane right ventricular contrast cineangiography for calculating right ventricular ejection fraction was good when all data points (n = 54) in all 28 subjects were compared (r = 0.93, standard error of the estimate [SEE] = 4.58 [p < 0.001], y = 1.0 X + 3.05). A comparison of resting and ejection fraction values determined at rest and during exercise revealed similar results (r = 0.91, SEE = 4.01 [p < 0.001], Y = 0.95 x + 4.8 for rest studies; r = 0.93, SEE = 5.23 [p < 0.001], Y = 1.01 x + 2.95 for exercise studies) (Fig. 1). Similar results were also observed when rest and exercise ejection fraction measurements were analyzed separately in normal subjects and patients. In normal subjects, the correlation obtained at rest was r = 0.74, SEE = 5.31 (p < 0.02), whereas during exercise the correlation wasr = 0.75, SEE = 4.7 (p < 0.02) (Fig. 2A). When rest and exercise data in normal subjects are considered together, the regression line is described by r = 0.84, SEE = 4.85 (p < 0.001), y = 0.95 x + 5.63 (Fig. 2A). When both imaging methods were compared in patients in Group II (Fig. 2B), the correlation coefficient of right ventricular ejection fraction was higher in the studies at rest (r = 0.92, SEE = 3.45 [p < 0.001]) than in the exercise studies (r = 0.9, SEE = 5.72 [p < 0.00 I]). Comparison of both methods with respect to combined rest and exercise data points (n = 34) resulted in values of r = 0.91, SEE = 4.56 (p < 0.001). The related linear regression equation is y = 1.02 x + 2.1 I . The correlation in Group II was better than that in Group I

NIENABER ET AL. RIGHT VENTRICULAR IMAGING WITH KRYPTON-81m

l ACC Vol. 5, No. 3 March 1985:687- 98

691

Table 2. Exercise Protocol and Hemodynamic Data During Krypton-81m Radionuclide Ventriculography and Cineangiography Cine angiograp hic Exercise Studies

Krypton-81 m Radionucl ide Exe rcise Studies

HR (beats/min)

SBP (mm Hg)

Re

Re

Ex

Ex

RP P (HR x SBP x 10") Re

Ex

Work Load (kpm/ min)

T ime (min)

HR (beats/min)

SBP (mm Hg)

Re

Re

Gro up I: Normal Sub jects (n Mean value :t SD p (Re vs . Ex )

74 117 :t 14 :t 13 p < 0 .001

125 166 :t 15 :t 16 p < 0 .001

9 .3 19 .4 :t 1.6 :t2.6 P < 0 .001

300

9

=

Ex

Ex

RPP (HR x S BP x 10 ' ) Re

Ex

Work Load T ime (kpm/ rnin) (min)

10)

120 72 ± 16 ± 16 P < 0 .001

12 1 163 ± 17 1: 17 P < 0 .01

19 .6 8.7 ± 1.9 ± 2 .1 P < 0.00 1

300

9

20 . 1 6 .6 21.7 9 .4 10.4 19 .5 10.4 20 .8 8.8 2 1.8 26 .4 8 .6 10 .1 25 .5 25 .7 8 .8 10.4 11.0 22 .7 9.4 ± 1.3 ± 2.8 P < 0 .00 1 9 .2 20 .7 9 .8 23.8 24 .7 8.8 24.4 11.4 9 .8 23 .4 ± 1.2 :t 1.9 P < 0 .0 1 19 .4 7 .4 22 .8 8. 1 9 .9 18 .6 9 .6 19 .7 8 .8 20. 1 ± 1.2 ± 1.9 P < 0 .0 1

300 300 300 150 60 300 300 300

6 6 6 4 4 6 9 8

25 1 1: 93

6 .3 ± 1.8

300 300 300 300 300

9 9 9 9 9

300 300 150 300 262

9 9 9 9 9

Gro up II: 18 Patient s With Right Ventr icular Abnormalities Case (no .)

I

62 126 118 84 84 122 138 88 148 96 78 130 134 86 72 128 100 86 13 1 83 ± 11 ±9 p < 0. 001 84 134 142 82 136 72 152 90 141t 82 ±8 ±9 ±SD P (Re vs. Ex ) p < 0 .001 114 15 60 108 16 58 17 84 130 114 18 78 Mean value 117 70 ± SD ± 13 ± IO p (Re vs . Ex) p < 0.01

2 3 4 5 6 7 8 9 10 Mean value :tSD p (Re vs. Ex) II 12 13 14 Mean value

115 150 120 170 130 170 110 145 105 140 115 190 100 150 125 205 105 115 114 165 ± IO ±23 P < 0 .001 110 145 125 170 130 195 130 170 124 170 :t 10 :t 2 1 P < 0 .01 130 170 140 195 115 150 120 165 126 169 ±11 ± 19 P < 0 .0 1

18.9 7. 1 10 .1 20 .1 20.7 10.9 9.7 20 .0 20 .7 10.1 24 .7 9 .0 20 . 1 8 .6 26 .2 9 .0 10.5 9 .9 2 1.4 9 .5 :t 1. 1 ± 2 .6 P < 0 .00 1 9 .2 19 .4 24. 1 10.3 26. 5 9 .4 11.7 25 .8 23 .9 10.1 ± 1.2 ±3 .2 P < 0 .01 7 .8 19.4 21.1 8.1 19 .5 9.7 9.4 18.8 19.7 8. 8 ± 1.0 ± 1.0 P < 0 .0 1

*p < 0 .05 ; t p < 0 .01 versus norm al subjects . Ex pressure; SD = standard dev iation.

=

300 300 300 150 60 300 300 300

7 6 6 6 6 6 9 10

25 1 1: 93

6 .9 :t 1.6

300 300 300 300 300

9 9 9 9 9

300 300 150 300 262

9 9 9 9 9

exercise; HR

since it encompassed a wider range of ejection fraction values. Linear regression curves in all sets of correlations revealed slightly higher ejection fraction values with the countbased scintigraphic method as compared with the cineangiographic method based on right ventricular volume estimations determined according to Simpson' s rule (Fig. I and 2). Because the majority of studies were performed both at rest and during exercise, we correlated the change in right ventricular ejection fraction during exercise obtained by both methods (Fig. 3). This correlation is r = 0.73 , SEE = 6.1 (p < 0 .001). The linear regression equation y = 0.78 x

=

130 60 82 124 80 118 90 134 88 150 132 72 150 88 61l 132 104 81l 134* 82 ± 13 ± II P < 0 .00 1 84 138 140 82 130 61l 81l 148 139t 80 :t9 ±8 P < 0 .00 1 58 118 62 114 120 86 80 116 71 117 ± 14 ±3 P < 0 .0 1

heart rate; Re

=

lID 155 105 175 130 165 liS 155 100 145 120 200 105 170 130 195 100 125 114 170 :t 12 :!:20 P < 0 .00 1 110 150 120 170 130 190 130 165 123 172 ± 7 :!:20 P < 0 01 128 164 130 200 115 155 120 170 123 172 ± 7 :!:20 P < 0 .0 1

rest; RPP

=

rate -pr essure produ ct ; SBP

=

systolic blood

+ ] .93 describing this relation is close to the line of identity and separates subjects with normal and abnormal functional right ventricular reserve . Since 9 of the IO normal subjects had an increase in right ventricular ejection fraction greater than 5% (ejection fraction units) during submaxirnal exercise, a decrease or no change in ejection fraction was considered a disturbed functional reserve. Inter- and intraobse rver variability. The interobserver variability for determination of right ventricular ejection fraction from krypton-Sl m equilibrium blood pool images averaged 5 .2% (ejection fraction units) for studies at rest and 5.5% for studies during exercise . For cineangiographic

lACC Vol. 5. No.3

NIENABER ET AL. RIGHT VENTRICULAR IMAGING WITH KRYPTON-81m

692

March 1985:687-98

GROUP 1+ 11

right ventricular ejection fraction, similar values of interobserver variability were found: 4.4% for studies at rest and 6.0% for studies during exercise. The intraobserver variability for krypton-81m studies was of comparable magnitude, averaging 5.0% for rest studies and 5.9% for exercise studies; these values did not differ from cineangiographic values for intraobserver variability in determining right ventricular ejection fraction at rest and exercise, that is, 4.5 and 5.8%, respectively.

Kr81m RVEF [%)

o

• REST

70

o ExERCISE

60 50

0

.,

40

.

v'

SEE

n-

20

~

SEE · 4 01 n · 28 Cres,1

n

"

v » 1.01 . · I

·

SEE' n ·

10

Right ventricular ejection fraction from gated equilibrium krypton-81m blood pool scans in normal subjects (Fig. 4A). At rest, right ventricular ejection fraction

~

v 09S •• 4 ti , . 0 9 11p,0001l

hi

30

1 0 . · 3 0S 09 3 '0 <0 00 1' 4 !Xl

20

30

40

50

averaged 52.7 ± 7.4% and was not statistically different from cineangiographic right ventricular ejection fraction (50.4 ± 6.2%). With submaximal dynamic exercise, right ventricular ejection fraction increased by 11.0% to 63.7 ± 6.7% (p < 0.001 versus rest).

2 ,9 ~

0 .93 IP< O001 1

on 26

60

( @ . pr CI ~ 1

70 [%)

Simpson's RVEF

Right ventricular function in pulmonary hypertension (Fig. 4B). In the nine patients with secondary pulmonary

Figure I. Comparison of right ventricular ejectionfraction (RVEF) in 28 subjects, including 10 normal subjects and 18 patients with altered right ventricularfunction or geometry, obtained by krypton81m multiple gated equilibrium blood pool scans and biplane right ventricular angiography (Simpson's rule). Separate linear regression lines are given a) for all data points (n = 54) from all subjects at rest (closed circles) and during exercise (open circles), y = 1.0 x + 3.05, r = 0.93 (p < 0.001), SEE = 4.58; b) for all measurements at rest (n = 28), y = 0.95 x + 4.8, r = 0.91 (p < 0.(01), SEE = 4.01; and c) for exercise data in all subjects (n = 26), Y = 1.01 x + 2.95, r = 0.93 (p < 0.(01), SEE = 5.23.

A

B

GROUP I

GROUP II

Kr81m RVEF

Kr81m RVEF

[%]

[%) • REST

• REST

70

70

o ExERCISE

•

60

,/

60

/' .·.1

50

/ '

v · 0.9S.· S6 3

30

,,0 ,,-

,/

/ '

r · 084 ID<0 .00 1' ~E ' . ~ n - 10

bl

/ n

20 ,...;/.

v · 0 ,89 .

8 04

SEE· S 3 1 n -

cl

20

30

40

50

v » 079 .. • 16,0

60

o v » 1.02. · 2J 1 • • 0.91 10<0.0011

40

SEE' ' .$6 n ' 34 bl

30

• • 0.92 '0 <0.0011

20

n·

10 (@.e'f CI1le)

70 [%]

Simpson's RVEF

v . 0 .94 ... 4 .8

SEE' 3.4S n · 18 h Mt l cI v · 1.06 1 · 0 .71 r · 0 .9 lp <0 00 1) SEE' S.12

IOlrnll

, . 0 .7510...0 .021 SEE' 4.1 n -

10

+

, . 0 .7410< 0 .02)

/ " b

0 EXERCISE

o

/.

"..,/(/

50 40

hypertension and the one patient with primary pulmonary hypertension, the average right ventricular ejection fraction at rest was 42.5 ± 5.0%, which differs from that in control subjects at rest (p < 0.05). In this subgroup of patients, however, no significant change in ejection fraction was found with dynamic exercise (40.6 ± 8.7%, P = NS). In four patients (Patients 1, 4, 6 and 7), ejection fraction even decreased with exercise in association with a marked in-

10

20

30

40

50

60

16 le lHCI1.e1

70 [%]

Simpson's RVEF

Figure 2. Comparison of right ventricular ejection fraction (RVEF) measurements obtained by krypton81m multiple gated equilibrium scintigraphy and biplane angiography using Simpson's rule. A, Separate regression lines are calculated a) for all data points (n = 20) from IO normal subjects at rest (closed circles) and during exercise (open circles), y = 0.95 x + 5.63, r = 0.84 (p < 0.(01), SEE = 4.85; b) for all measurements at rest (n = 10), Y = 0.89 x + 8.04, r = 0.74 (p < 0.02), SEE = 5.31; and c) for exercise data (n = 10), Y = 0.79x + 16.0, r = 0.75 (p < 0.02), SEE = 4.7. B, Eighteen patients with functional or geometric right ventricular abnormalities. Separate linear regression lines are given a) for both rest and exercise right ventricular ejection fraction (n = 34), y = 1.02 x + 2.11, r = 0.91 (p < 0.001), SEE = 4.56; b) for all measurements at rest (n = 18), Y = 0.94x + 4.8, r = 0.92 (p < 0.(01), SEE = 3.45 16), and c) for exercise data (n y = 1.06x + 0.71, r = 0.9 (p < 0.(01), SEE = 5.72.

NIENABER ET AL. RIGHT VENTRICULAR IMAGING WITH KRYPTON-81m

JACC Vol. 5. No.3 March 1985:687-98

ventricular ejection fraction at rest was 14%. Patients 9 and 10 did not perform exercise since both were symptomatic at rest (New York Heart Association functional class IV).

Kr81m ~RVEF [%] 25

+ 1.9

y = 0.78 x

Right ventricular function in right ventricular outflow tract obstruction (Fig. 4C). Average ejection fraction in

= 0.73 Ip < O.OOli

r

SEE = 6.1 n =

.. / ">"

20

26

15 10 -

-15

four patients with altered right ventricular anatomy as a consequence of chronic right ventricular outflow tract obstruction (tetralogy of Fallot and pulmonary stenosis) was found to be normal (48.25 ± 4.6%) . During exercise . there was a physiologic increase to 63.5 ± 7.1% (p < 0.01 versus rest). No patient in this subset was symptomatic; however, all four patients revealed marked right ventricular hypertrophy, right heart dilation and highly elevated systolic right ventricular pressure resulting in a mean valvulopulmonary pressure gradient of 55.7 ± 22.1 mm Hg (range 36 to 92) at rest and 62.5 ± 22.3 mm Hg (range 37 to 89) during exercise. Pulmonary artery pressure both at rest (16 ± 3.6 mm Hg) and during exercise (22 ± 3 mm Hg) did not differ from that in normal subjects (Table I). However, in patients with pulmonary hypertension during exercise pulmonary artery pressure was 49 ± 7 mm Hg (p < 0.001). Pulmonary capillary wedge pressure both at rest and during exercise was normal (7.3 ± I and 9.7 ± 2 mm Hg, respectively), but right atrial pressure increased from 8 ± 2 mm Hg at rest to 12 ± 2 mm Hg during exercise (p < 0.05), which was higher than normal (p < 0.01) . Cardiac output . however , was normal at rest (4 .7 ± 0.5 liters/min) and during exercise (9.6 ± 2.4 liters/min) (Table I).

~

5/

°

~

t

a

-10

I

-5

/

/0

5

t:.

I

I

15

20

Simpson's ~RVEF[%1

-10

°

0

I

10

-5

0

-15

o PH ... LV dysfunction • TOF + PS A INF + RVM I

• normal subjects

Figure 3. Comparison of exercise-induced changes in right ventricular ejection fraction (6 RVEF) obtained by krypton-81 mequilibrium blood pool scans and biplane right ventricular cineangiography in 10normal subjects and 16 patients with right ventricular abnormalities: pulmonary hypertension (PH) and left ventricular (LV) dysfunction in 8, right ventricular outflow tract obstruction (tetralogy of Fallot [TOF] and pulmonary stenosis [PS]) in 4 and inferior wall and rightventricular infarction (lNF + RVMI) in 4. y = 0.78 x + 1.9, r = 0.73 (p < 0.(01), SEE = 6.1.

Right ventricular function in right ventricular infarction (Fig. 4D). Complete obstruction of the proximal

crease in right ventricular afterload reflected as an exerciseinduced increase in pulmonary resistance. There was a marked increase in mean pulmonary artery pressure from a baseline value of 36 ± 19 to 49 ± 7 mm Hg (p < 0.01) and an inadequate increase in cardiac output from 4.4 ± 1.3 to 6.5 ± 1.6 liters/min (p < 0.01) (Table 1), whereas heart rate increased from 82 ± 13 to 134 ± II beats/min (p < 0.00 I) (Table 2). Mean right atrial pressure and the atrial V wave also increased on exercise from I I ± 5 to 17 ± 4 mm Hg (p < 0.01) and from 14 ± 5 to 26 ± 8 mm Hg (p < 0.01), respectively (Table 1). In Patient 10, right

A normal subjects

B

PH ... PPH

Kr81m RVEF

6.

c

right coronary artery resulted in a right ventricular ejection fraction at rest of 37.5 ± 5.4%, which was lower than normal (p < 0.05). Moreover , during exercise no significant functional increase was detected (38.75 ± 8.3%, p = NS). In contrast to normal subjects and those with compensated right ventricular outflow tract obstruction, pulmonary capillary wedge pressure was slightly elevated to 13 ± 2 mm Hg at rest (p < 0.05) and 15 ± 3 mm Hg during exercise (p < 0.05). Cardiac output was normal at rest (5.0 ± 0.5

o INF +RVMI

rOF+ PS

no t Inc l uCIed In m ea n \Id! U~

I

[96]

f~!

40

r--

_

20 t.

0

t,

p
rest

exercise

L-

rest

693

-n s . ---.J

exercise

L

p
rest

exercise

L.-

rest

n.s.-!

exercise

Figure 4. Krypton-81 m right ventricular ejection fraction (RVEF) at rest and in response to dynamic bicycle exercise in normal subjects; A, patients with secondary pulmonary hypertension (PH) and primary pulmonary hypertension (PPH). B, patients with right ventricular outflow tract obstruction due to residual pulmonary stenosis (PS) (three patients with corrected tetralogy of Fallot (TOF) with residual pulmonary stenosis andone patient with subvalvular pulmonary stenosis and patients with complete obstruction of the right coronary artery and right ventricular and inferior myocardial infarction (INF + RVMI). Subjects in A andC show a physiologic response to exercise, whereas those in Band Dhave adisturbed right ventricular functional reserve.

694

NIENABER ET AL. RIGHT VENTRICULAR IMAGING WITH KRYPTON·81m

lACC Vol. 5. No. 3 March 1985:687-98

liters/min) , but subnormal during exercise (8.5 liters/min) (p < 0 .01) (Table I) .

Discussion Correlation of radionuclide and cineangiographic assessment of right ventricular ejection fraction. This study describes and validates the clinical use of a new noninvasive scintigraphic method using ultrashort-lived krypton-81 m to determine ejection fraction from multiple gated equilibrium blood pool images of the isolated right heart chambers and the pulmonary artery . Special attention was given to dynamic right ventricular function and anatomy (Fig. 5). A high degree of correlation was found with cineangiographically determined right ventricular ejection fraction (Simpson 's rule based on geometric assumptions) both in normal subjects and patients with a variety of right ventricular abnormalities . Since most of the studies were done in the resting state and during dynamic submaximal bicycle exercise, an extended range of right ventricular ejection fraction measurements could be studied. In both normal subjects and those with heart disease, the two methods correlated well, with a tendency to slightly more variability during dynamic exercise in patients with right ventricular abnormalities. Whereas different levels of exercise in both studies in each subject were carefully avoided (Table 2), the reason for that increas ed variably with exercise could be a decreasing accuracy in defining cineangiographic ventricular contours and axes with rapid contrast washout. Although we extended the axes to the maximal even faintly opacified outline of the radiopaque projection (30), all formulas for calculating right ventricular volume were validated in rest studies (10-15) and then may be less useful for estimating right cavity volume during dynamic exercise. Count-based methods for assessing right ventricular function , however, are known to be relatively independent of geometric assumptions . With continuous infusion of krypton-81 m and equilibrium acquisition, the right-sided radioactivity is almost homogeneously distributed since the exch ange rate in the heart is higher than the half-life of krypton-81 m. Thus , even if streaming or incomplete mixing occurred, the calculation of ejection fraction using count density ratio methods remains valid. Moreover, unstable delivery of krypton-81m was excluded by recording stable baseline count rates. Advantages of imaging with krypton-81m compared with technetium-99m. Due to high photon flux, rapid decay and complete clearance of radioactivity on the alveolar surface, superimposition of right- with left-sided activity is negligible (9). These phy sical properties of krypton-81 m allow the use of multiple different projections for right heart imaging. In initial pilot trials , we found a variable 15 to 30° right anterior oblique view to provide the best separation between the right atrium and right ventricle, especially in

RVED frame with RVED ROI

RVES frame with RVED and RVES ROI Figure 5. Normal right ventricular function. Krypton-8l m gated blood pool scan in the right anterior oblique view showing norma l right heart structures: right ventricular end-diastolic (RVED) frame with right ventricular end-diastol ic region of interest (RO!) in the upper panel and the corresponding end-systolic (RVES ) frame with end-diastolic and end-sy stolic region of interest in the lower panel (right ventricular eject ion fraction = 54%). The 25° right anterior oblique view provides a clear outline of the right atrium (RA), right ventricle (RV) and pulmonary artery (PA) as well as demarcation of the swinging tricuspid valve (TV) and the relatively stable pulmonary valve (PV) plane.

patients with enlarged right heart dimensions due to chronic pressure or volume overload . No overlap even with a dilated atrium was encountered, and both the tricuspid and the pulmonary valve planes were clearly detectable without the help of phase images in the majority of our patients (Fig. 5, 6 and 7). Imaging with krypton-81 m provides insight into morphologic abnormalities that are not accessible with conven tional technetium-99m radiopharmaceutical agents using equilibrium mode acquisition in the 30 to 45° left anterior oblique view. This position, however, requires variable right ventricular regions of interest to minimize overlapping activity from the right atrium (20,22). In our study , separate end-diastolic and end-sy stolic regions of interest were drawn to compensate for extensi ve motion of the tricuspid valve plane (Fig. 5), especially in the six patients with associated tricuspid valve regurgitation (Fig . 7). Limitations of imaging with krypton 81-m. Right ventricular volume determination is not included in our analysis of right ventricular function. Volume measurements would

NIENABER ET AL. RIGHT VENTRICULAR IMAGING WITH KRYPTON-81m

lACC Vol. 5. No.3

March 1985:687-98

695

RVED frame

end-diastolic frame

RVES frame

end-systolic frame

Figure 6. Patient 4. Krypton-81 m gated blood pool scan in right ventricular end-diastolic (RVED) and end-systolic (RVES) frame in the 30° right ventricular oblique projection in Patient 4 with secondary pulmonary hypertension, chronic obstructive pulmonary disease and associated tricuspid regurgitation . Right ventricular ejection fraction at rest is 44%. The right atrium (RA) is enlarged . Other abbreviations as in Figure 5.

Figure 7. Patient 10. Krypton -81m right ventricular gated blood pool scan in Patient 10 with primary pulmonary hypertension . The image shows enlarged right atrial (RA) and right ventricular (RV) cavities, with minimal contractile motion due to highly elevated pulmonary resistance . Right ventricular ejection fraction at rest is 14%. The pulmonary valve (PV) is clearly defined , whereas the tricuspid valve (TV) is more difficult to recognize as a result of pressure and volume overload of the right heart chambers.

only be accessible by geometry-based calculation from single plane or biplane krypton-81 m images; the physical properties of krypton-81 m do not provide peripheral blood sampling for reference activity (31) such as with technetium99m (32). Conventional radionuclide methods thus far are critically dependent on the relative surface of the diastolic and systolic areas (20) or on the relative sampling of atrial and pulmonary structures (22) . Imaging with krypton-81 m does not necessitate any of these background corrections to achieve an excellent correlation (Fig. 1 and 2) . However, with extremely low ejection fraction values. the lung activity included in the end-diastolic region of interest may possibly increase and become significant. It is unknown whether this potential source of error is completely counterbalanced by the rapid decay of krypton-81 m. To date, the krypton-Sl m generator is not yet commercially available. Because of the half-life of the parent rubidium-81, its use is limited to 18 hours after charge . Radiation of krypton-81m. The radiation dose of krypton-81 m is calculated to be substantially lower as compared

tion is a well established method for estimating right ventricular volume and right ejection fraction (13,14,36) and is probably the most useful for validation . However, as an invasive procedure , it is not suitable for serial measurements and bears inherent difficulties related to the geometry of the right ventricle and complex mathematical formulas for volume calculations. Recentl y, several radionuclide techniques using either first pass (1,19 ,37) , gated first pass (38) or gated equilibrium acquisition mode (20 ,22) and even echocardiographic methods (39-42) have been used to " measure " right ventricular ejection fraction. The count-based radionuclide methods using technetium 99-m are less dependent on assumptions about complex geometry. A good

with conventional doses of technetium-99m or even repeated

correlation was found among these methods when patients

injections of short-lived gold-196m (33.34). While the activity delivered to the right ventricle was calculated to be 2

with coronary artery disease were included (17,22,42). Our data obtained from isolated right-sided equilibrium images

to 3 mCi/min, the radiation burden to the patient is almost insignificant due to short half-life and complete exhalation (35). Rubidium-81 breakthrough (half-life 4.7 hours) accounts for only 0.1 JLClml and is, therefore, not a quantitatively important source of radiation .

Comparison with results by other noninvasive methods. Contrast ventriculography during cardiac catheteriza-

696

NIENABER ET AL. RIGHT VENTRICULAR IMAGING WITH KRYPTON-81m

lACC Vol. 5, No.3 March 1985:687-98

using ultrashort-lived krypton-81m are almost identical to results by others (1,9,19,20,22,37,38,42) using the first pass and equilibrium methods both in normal subjects and patients with coronary artery disease.

pulmonary artery in the 30° right anterior oblique view. However, the normal pulmonary capillary wedge pressure suggests normal pulmonary vascular resistance (afterload) and normal left ventricular function (Table I).

Right ventricular function in pulmonary hypertension. Our results further indicate that right ventricular sys-

Right ventricular function in inferior myocardial infarction. The abnormal right ventricular ejection fraction

tolic function is frequently depressed in adults with chronic pulmonary hypertension (Fig. 4B). Although the left ventricle is not visible and accessible with intravenous krypton81m, it can be anticipated from the elevated left ventricular end-diastolic pressure in this subgroup of patients (Table I) that a depressed right ventricular function at rest and especially during exercise is caused by an increased afterload or systolic wall stress due to valvular heart disease or left ventricular dysfunction, or both (43-46); variables of right ventricular performance independent of afterload (that is, maximal rate of isovolumic pressure increase) are normal in pulmonary hypertension (37,44). Depressed right ventricular systolic function in this group of patients probably results from abnormally high wall stress according to the law of Laplace rather than from decreased contractility (19,47-50), resulting in tricuspid regurgitation, particularly in relation to exercise, and a right atrial V wave of 26 ± 8 mm Hg (p < 0.01). Forward cardiac output is severely depressed. Role of tricuspid regurgitation. In six patients with globally compromised right and left ventricular function after both mitral and aortic valve replacement, functional (or dilative) tricuspid regurgitation was observed (Patients 1,2,4,5,6 and 7, Table I). Under resting conditions, an almost normal ejection fraction was found despite a low (anterograde) output state (Fig. 6). With dynamic exercise, however, the right ventricle dilates even more and ejection fraction decreases as pulmonary hypertension (afterload) and venous return (preload) increase (51). This indicates that in these patients, the abnormal right ventricular reserve may be a physiologic response to augmented loading rather than evidence for intrinsic right ventricular dysfunction.

in patients with inferior myocardial infarction involving parts of the right ventricle (Fig. 4B) is probably not a reflection of an altered loading condition since the pulmonary artery pressures are normal. However, it does suggest regional alteration in the right ventricular contraction pattern (57,58). The mildly elevated left ventricular end-diastolic pressure is probably the result of additional inferior wall infarction. From a methodologic point of view, the analysis of regional wall and septal motion is feasible, especially with variable camera projections and with the help of phase and amplitude motion images. Further studies with emphasis on regional wall motion analysis are required to substantiate the use of krypton-Sl m ventriculography. Clinical applications. Blood pool imaging with krypton-81 m appears to be useful whenever serial nontraumatic assessment of right ventricular function is required. It can be sequentially applied in the coronary care unit, either for

Figure 8. Patient 13. Krypton-81 m right ventricular gated blood pool scan in Patient 13 with corrected tetralogy of Fallot and residual pulmonary stenosis (PS). Right ventricular ejection fraction is normal at rest (51%) and during physical exercise (69%). The anatomic site of the pulmonary stenosis can be located on the 30° right anterior oblique krypton-81 m image. Other abbreviations as in Figure 5.

Right ventricular function in right ventricular outflow tract obstruction. The degree of right ventricular hypertrophy obviously accounts for much of the variability of right ventricular performance in the presence of pressure overload. Children with isolated pulmonary stenosis and severe right-sided hypertrophy have a normal or increased ejection fraction (12,16). This was observed in our patients with residual pulmonary stenosis after surgical repair of tetralogy of Fallot and organic pulmonary stenosis, respectively (Fig. 4C). In contrast to the results of Reduto et al. (52), both rest right ventricular ejection fraction and functional reserve were normal, but right ventricular volume was elevated; thus, right ventricular dysfunction is obviously prevented by compensatory chronic hypertrophy (53-56), resulting in normal systolic wall stress and normal ejection fraction. Figure 8 clearly demonstrates the dimensional difference between the enlarged right cavity and the

end-diastolic frame

end-systolic frame

NIENABER ET AL. RIGHT VENTRICULAR IMAGING WITH KRYPTON-81m

lACC Vol. 5, No.3

March 1985:687-98

postoperative patients or in the setting of acute myocardial infarction or pulmonary embolism. In combination with physical exercise, it may be used to screen for right ventricular dysfunction and pulmonary hypertension. Moreover, since the generator can alternatively be used for lung ventilation scintigrams in conjunction with perfusion scans, it may be, despite its relatively high costs, of particular interest to a cardiopulmonary unit in cooperation with the nuclear medicine departments. We thank Manfred Prinz and Barbara Zink for their excellent graphic and technical assistance and Roswitha Hartfelder for preparing the manuscript and expert secretarial assistance.

References I. Yano Y, Anger HO. Ultrashort-lived radioisotopes for visualizing blood vessels and organs. 1 Nucl Med 1968;9:2-6.

2. Clark lC, Horlock PL, Watson IA. Krypton-81m generators. Radiochem Radioanal Let 1976;25:245-8. 3. Jones T, Clark lC. A cyclotron produced 8IRb-8Im'Kr generator and its uses in gamma camera studies (abstr). Br 1 Radiol 1969;42:237-41. 4. Yano Y, McRae 1, Anger HO. Lung function studies using shortlived radionuclides and the scintillation camera. 1 Nucl Med 1970;11:764-9. 5. Arnot RN, Glass HI, Clark lC. Radioaktive Isotope. In: Hofer R, ed. Klinik and Forschung. vol 9. Munchen: Urban-Schwarzenberg, 1970:76-8. 6. Selwyn AP, Jones T, Turner JH, Pratt T, Clark 1, Lavender P. Continuous assessment of regional myocardial perfusion in dogs using krypton-81m. Circ Res 1978;42:771-80. 7. Selwyn AP, Forese G, Fox K, Jonathan A, Steiner R. Patterns of disturbed myocardial perfusion in patients with coronary artery disease: regional myocardial perfusion in angina pectoris. Circulation 1981;64:83-90.

697

19. Berger Hl, Matthay RA, Loke 1, Marshall RC, Gottschalk A, Zaret BL. Assessment of cardiac performance with quantitative radionuclide angiography: right ventricular ejection fraction with reference to findings in chronic obstructive pulmonary disease. Am 1 Cardiol 1978;41:897-901. 20. Maddahi 1, Berman DS, Matsuoka DT, et al. A new technique for assessment of right ventricular ejection fraction using rapid multiple gated equilibrium cardiac blood pool scintigraphy. Description, validation and findings in chronic coronary artery disease. Circulation 1979;60:581-90. 21. Borer IS, Bacharach SL, Green MV, Kent KM, Epstein SE, Johnson GS. Real-time radionuclide cineangiography in the noninvasive evaluation of global and regional left ventricular dysfunction at rest and during exercise in patients with coronary artery disease. N Eng 1 Med 1977;296:839-44. 22. Slutsky R, Hooper W, Gerber R, et al. Assessment of right ventricular function at rest and during exercise in patients with coronary heart disease: a new approach using equilibrium radionuclide angiography. Am 1 Cardiol 1980;45:63-71. 23. Colobetti LG, Mayron LW, Kaplan E, et al. Continuous radionuclide generation. I. Production and evaluation of a krypton-81m minigenerator. 1 Nucl Med 1974;15:868-74. 24. Mayron LW, Friedman AM, Kaplan E, etal. A sterile multicomponent 81 Rb-8lm Kr minigenerator and delivery system. Int 1 Nucl Med Bioi 1975;2:141-5. 25. Kaplan E, Mayron LN. Evaluation of perfusion with the 81 Rb-8lm Kr generator. Semin Nucl Med 1976;6: 163-7. 26. lones T, Clark lC, Hughes 1M, et al. 8lm-Kr generator and its uses in cardiopulmonary studies with the scintilation camera. 1 Nucl Med 1970;11:118-24. 27. Clark JC, Horlock PL, Watson IA. Krypton-81m generators. Radiochem Radioanal Let 1976;25:245-54. 28. Chan WWC, Kalff V, MacDonald D II, et al. Topography of preemptying ventricular segments in patients with Wolff-Parkinson-White syndrome using scintigraphic phase mapping and esophageal pacing. Circulation 1983;67:1139-46. 29. Botvinik EH, Frais MA, Shosa DW, et al. An accurate means of detecting and characterizing abnormal pattern of ventricular activation by phase image analysis. Am 1 Cardiol 1982;50:280-5.

8. Selwyn AP, Steiner R, Kivisaari H, Fox K, Forese G. Krypton-81m in the physiologic assessment of coronary arterial stenosis in man. Am 1 Cardiol 1979;43:547-53.

30. Sandler H. Dimension analysis of the heart. A review. Am 1 Med 1970;260:56-61 .

9. Knapp WH, Helms F, Lambrecht RM, Elfner R, Gasper H, Vollhaser HH. Kr-81m for determination of right ventricular ejection fraction. Eur 1 Nucl Med 1980;5:487-92.

31. Nickel 0, Schad N, Andrews El, Fleming rw, Mello M. Scintigraphic measurements of left ventricular volumes from the count-density distribution. 1 Nucl Med 1982;23:404-10.

10. Goerke Rl, Carlsson E. Calculation of right and left cardiac ventricular volumes. Invest Radiol 1967;2:360-5.

32. Sorensen SG, Ritchie Jl., Caldwell JH, Hamilton GW, Kennedy lW. Serial exercise radionuclide angiography: validation of count-derived changes in cardiac output and quantitation of maximal exercise ventricular volume changes after nitroglycerin and propranolol in normal men. Circulation 1980;61:600-9.

II. Arcilla RA, Tsai P, Thilenius 0, Ranninger K. Angiographic method for volume estimation of right and left ventricles. Chest 1971;60:446-50. 12. Graham TP, larmakani 1M, Atwood GF, Canent RV. Right ventricular volume determination in children. Circulation 1973;47: 144-9. 13. Gentzler RD, Brisilli MF, Gault JH. Angiographic estimation of right ventricular volume in man. Circulation 1974;50:324-8. 14. Ferlinz 1, Gorlin R, Cohn PF, Herman MW. Right ventricular performance in patients with coronary artery disease. Circulation 1975;52:608-15. 15. Thilenius OG, Arcilla RA. Angiographic right and left ventricular volume determination in normal infants and children. Pediatr Res 1974;8:67-73.

33. Dymond DS, Elliot AT, Flatman W, et al. The clinical validation of gold-195m: a new short half-life radiopharmaceutical for rapid sequential, first-pass angiography in patients. 1 Am Coil Cardiol 1983;2:85-92. 34. Ackers lG, de long RB1. Dosimetry consequence of eluates from Hg195m/Au-195m generators (abstr). 1 Nucl Med 1982;23:68. 35. Goris M, Daspit T. Krypton-81m. In: Guter M, ed. Progress in Nuclear Medicine. vol. 5. New Radiogases in Practice. New York: S. Karger, 1978:209-28.

16. Fisher EA, Dubrow IW, Hasterreiter AR. Right ventricular volume in congenital heart disease. Am 1 Cardiol 1975;36:67-75.

36. Chapman CB, Baker 0, Reynolds 1, et al. Use of biplane cinefluorography for measurements of ventricular volume. Circulation 1958;18:1105-9.

17. Steele P, Kirch D, LeFree M, Dennis B. Measurement of right and left ventricular ejection fraction by radionuclide angiography and coronary artery disease. Chest 1976;70:51-9.

37. lohnson LL, McCarthy DM, Sciaca RR, Cannon Pl. Right ventricular ejection fraction during exercise in patients with coronary artery disease. Circulation 1979;60: 1284-91.

18. Schuler G, Petersen K, Ashburn W, et al. Right ventricular performance following surgical relief of pulmonary hypertension. 1 Nucl Med 1978;19:736-44.

38. Dehmer Gl, Firth BG, Hillis LD, Nicod P, Willerson rr, Lewis SE. Nongeometric determination of right ventricular volumes from equilibrium blood pool scans. Am 1 Cardiol 1982;49:78-84.

698

NIENABER ET AL. RIGHT VENTRICULAR IMAGING WITH KRYPTON-81m

JACC Vol. 5, No.3 March 1985:687-98

39. Bommer W, Weinert L, Neumann A, Neffe J, Mason DT, DeMaria A. Determination of right atrial and right ventricular size by twodimensional echocardiography. Circulation 1979;60:91-100.

49. Matthay RA, Berger HJ, Davies RA, et al. Right and left ventricular exercise performance in chronic obstructive pulmonary disease: radionuclide assessment. Ann Intern Med 1980;93:234-9.

40. Limacher MC, Quinones MA, Poliner LR, Nelson JG, Winters WL, Waggoner AD. Detection of coronary artery disease with exercise two-dimensional echocardiography. Description of a clinically applicable method and comparison with radionuclide ventriculography. Circulation 1983;67: 1211-8.

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

41. Zwehl W, Gueret P, Meerbaum 5, Holt D, Corday E. Quantitative two-dimensional echocardiography during bicycle exercise in normal subjects. Am J Cardiol 1981;47:865-73. 42. Panidis IP, Ren J-F, Kotler MN, et al. Two-dimensional echocardiographic estimation of right ventricular ejection fraction in patients with coronary artery disease. J Am Coli Cardiol 1983;2:911-8. 43. Borer JS, Bacharach SL, Green MV, et al. Left ventricular function in aortic stenosis: a response to exercise and effects of operation. Am J Cardiol 1978;41:382-6. 44. Borer JS, Gottdiener JS, Rosing DR, et al. Left ventricular function in mitral regurgitation: determination during exercise (abstr). Circulation 1979;59:60(suppl II):II-38. 45. Borer JS, Bacharach SL, Green MV, et al. Exercise-induced left ventricular dysfunction in symptomatic and asymptomatic patients with aortic regurgitation: assessment with radionuclide cineangiography. Am J Cardiol 1978;42:341-57. 46. Slutsky R, Ackermann W, Hooper W, et al. The response of left ventricular ejection fraction and volume to supine exercise in patients with severe COPD (abstr). Circulation 1979;59,60(suppl II):II-234. 47. Rushmer R. Functional anatomy and control of the heart. In: Rushmer R, ed. Cardiovascular Dynamics, 4th ed. Philadelphia: WB Saunders, 1976:89-98. 48. Jezek V, Schrigen F, Sadoul P. Right ventricular function and pulmonary hemodynamics during exercise in patients with chronic obstructive bronchopulmonary disease. Cardiology 1973;58:20-31.

51. Stein PD, Sabbah HN, Anbe DT, Marzilli M. Performance of the failing and nonfailing right ventricle of patients with pulmonary hypertension. Am J Cardiol 1979;44:1050-5. 52. Reduto LA, Berger HJ, Johnstone DE, et al. Radionuclide assessment of right and left ventricular exercise reserve after total correction of tetralogy of Fallot. Am J Cardiol 1980;45: 1013-8. 53. Ruzyllo W, Nihill MR, Mullins CE, McNamara DG. Hemodynamic evaluation of 221 patients after intracardiac repair of tetralogy of Fallot. Am J Cardiol 1974;34:565-70. 54. Hawe A, McGoon DC, Kincaid DW, Ritter DG. Fate of outflow tract in tetralogy of Fallot. Ann Thorac Surg 1982;13:137-41. 55. Halm JR, Bowman Fa, Jaoneson AG, Ellis K, Griffith SP, Blumenthal S. An evaluation of total correction of tetralogy of Fallot. Circulation 1963;27:805-10. 56. Oku H, Shirotani H, Yokoyama T, et al. Right ventricular outflow tract prosthesis in total correction of tetralogy of Fallot. Circulation 1980;62:604-9. 57. Berger HJ, Johnstone DE, Sands JM, Gottschalk A, Zaret BL. Response of right ventricular ejection fraction to upright exercise in coronary artery disease. Circulation 1979;60:1292-9. 58. Bodenheimer MM, Banka VS, Foshee CM, Helfant RH. Simultaneous right and left ventricular performance in coronary heart disease: disassociation of ejection fraction and asynergy. Clin Res 1978;26: 220-6.