- Email: [email protected]
Gated radionuclide angiographic evaluation of valve regurgitation
Gated Radionuclide Angiographic Evaluation of Valve Regurgitation
EDWARD
W. BOUGH, MD, FACC
ELAN J. GANDSMAN,
PhD
DAVID L. NORTH, MS RICHARD S. SHULMAN,
MD, FACC
Providence, Rhode Island
From the Departments of Cardiology and Nuclear Medicine, The Miriam Hospital and Brown University, Providence, Rhode Island. Manuscript received January, 29, 1980; revised manuscript received April 14, 1980, accepted April 18, 1980. Address for reprints: Edward W. Bough, MD, Department of Cardiology, The Miriam Hospital, 184 Summit Avenue, Providence, Rhode Island 02906.
Gated radionuclide angiography is a new noninvasive technique that can be used to calculate the ratio of lefl and right ventricular stroke volumes. This stroke volume ratio, which must be unity in normal subjects,increases in patients with aortlc or mitral regurgitation in’direct proportion to the degree of left ventricular volume overload, provided no shunts or regurgitant right heart lesions are present. In 22 patients with aortic or mitral regurgitation there was excellent correlation between the stroke volume ratio determined with gated radlonuclide angiography and with standard quantitative catheterization methods (r = 0.79). Measurement of valve regurgitation with this radlonucllde method also correlated well with data obtained from semlquantltatlve aortic root or left ventricular clneanglography (r = 0.72). Twenty-one of the 22 patients with valve regurgitation had an abnormally elevated stroke volume ratio, thereby suggesting that gated radlonuclide angiography may be useful in detecting or excluding hemodynamically significant valve regurgitation.
Semiquantitative aortic root or left ventricular contrast cineangiography is the most common method of evaluating valve regurgitation.1-3 Combined quantitative ventriculography and cardiac output determinations allow more accurate quantitation of such regurgitation, but this technique is more complex and the results are less reproducible.4 Previously available noninvasive methods have been inadequate for accurate quantitation of valve regurgitation. Although M mode echocardiography is invaluable for demonstrating structural valve abnormalities, it allows only indirect inferences as to the severity of regurgitant lesions.5 However, pulsed Doppler echocardiography has recently been shown to be useful for localizing regurgitant lesions and for quantitating mitral regurgitation.6*7 Recent studiessyg using methods similar to ours have suggested that gated radionuclide angiography is a superior noninvasive technique for quantitating valve regurgitation. In this study we compare results of gated radionuclide angiography with those of both quantitative and semiquantitative cardiac catheterization methods in evaluating valvular heart disease in 40 patients. Methods Patients: Between November 1978 and June 1979, forty patients underwent gated radionuclide angiography within 48 hours of diagnostic right and left heart catheterization. Eighteen patients (14 male and 4 female) who were being evaluated for coronary artery disease had no angiographic or clinical evidence of valve regurgitation or shunts and served as a control group. Twenty-two pa-
September 1980
The American Journal of CARDIOLOGY
Volume 48
423
RADIONUCLIDE ANGIOGRAPHY IN VALVE DISEASE-BOUGH
ET AL
tients had typical clinical findings of aortic (n = 3), mitral (n = 10) or combined (n = 9) valve regurgitation that were confirmed at cardiac catheterization; none had clinical or catheterization evidence of tricuspid regurgitation (Table I). Cardiac catheterization and angiography: All patients underwent right and left heart catheterization in a fasting, mildly sedated state. Cardiac output and effective forward stroke volume were determined from multiple indicatordilution curves and simultaneously recorded heart rate. Quantitative single plane left ventricular cineangiograms were performed in the right anterior oblique projection and analyzed for ventricular volumes and ejection fraction in a standard mariner.... Aortic root cineangiography was performed in all patients with rheumatic valve disease and in other patients with clinical evidence of aortic regurgitation. In the absence of right heart regurgitant valve lesions anti shunts, it can be assumed that the right ventricular stroke volume (RVSV) is equal to the effective forward stroke volume (SV) determined from the cardiac output. Because the total left ventricular stroke volume (LVSV) was independently measured from quantitative cineangiographic data, the ratio of left to right ventricular stroke volumes, or stroke volume ratio (SVR), could be calculated as follows: LVSV (angiographic)
Sjq&!E= RVSV
Forward
SV (cardiac
output)
The stroke volume ratio directly quantitates total left ventricular volume overload whether one or both left heart valves are regurgitant. The regurgitant fraction (RF) of the left ventricular stroke volume has also been used as a quantitative index of valve regurgitation and was calculated from the same data as follows:* RP = LVSV - RVSV 1*x
SVR - 1
-
LVSV
-
SVR
Aortic root and left ventricular cineangiograms were graded for the severity of aortic or mitral regurgitation by two independent observers who did not know the patients’ iden-
tity. Each valve lesion was graded from l+ to 4+ according to commonly used scales; the sum for both left heart vaives was designated as the angiographic score.‘l.12 Grading disagreements between observers were resolved by simultaneous review of the films in question. Radionuclide angiography: All patients underwent gated radionuclide angiography after in vivo red cell labeling with technetium-99m pyrophosphate.‘:’ Cardiac imaging was performed with a gamma camera (Elscint LF) and a 30’ caudal slant hold collimator of intermediate sensitivity (Engineering Dynamics Corporation) in the left anterior oblique view that provided optimal ventricular separation. Gated images were collected with a computer (Digital Equipment Corporation, PDP 11/40) and an electrocardiographic synchronizer (Brattle Instruments) in a 64 by 64 matrix at a rate of at least 1S frames/s to a total of 1 million counts/frame; the total acquisition time was usually about 10 minutes. All radionuclide studies were analyzed by an operator who was unaware of the catheterization results. End-diastolic and end-systolic frames were determined from the time-activity curve of an approximated left ventricular region of interest and were used to generate two functional images by computer frame arithmetic. Subtraction of the end-diastolic from the end-systolic frame produced a ventricular stroke volume image in which each pixel displays counts proportional to ventricular volume change. Because maximal atria1 volume occurs at ventricular end-systole and minimal atria1 volume at ventricular end-diastole, “reverse” subtraction of the end-systolic from the end-diastolic frame produced an atria1 volume change image analogous to the ventricular stroke volume image.14 Computer superimposition of these two images results in a composite functional image with excellent delineation of atrioventricular borders and the ventricular regions, especially the right ventricular conus (Fig. 1). Although the composite functional image is generated from the end-diastolic and end-systolic frames of the left ventricular time-activity curve and is strictly accurate only for the left ventricular stroke volume, it provides an excellent approxi-
TABLE I Angiographic, Radionuclide and Catheterization Data in 22 Patients With Valve Regurgitation Gated Radionuclide
Angiography Lesion Grade
1+ MR l+ l-l- MR
1 1
2+ l+ MR 2+ MR 2+ MR 2+ MR 3+ MR
: 2 2
3+ 4+ MR AR 3+ AR 3+ AR 1+ MR, 2+MR, l+ MR, l+ MR, 2+ MR, l+ MR, 2+ MR, 2+ MR, 3+ MR,
:
AR = aortic regurgitation;
424
Score
September 1980
;
: 2+ AR l+AR 3+ AR 3f AR 2+ AR 4+ AR 3+ AR 3+ AR 3+ AR MR = mitral regurgitation:
3” 4 4 4 5 2 6 RF = regurgitant
The American Journal of CARDIOLOGY
SW? 1.35 1.60 1.13 1.27 1.44 1.31 1.43 1.25 1.33 2.83 i .3a 1.96 1.23 1.47 2.08 2.89 3.15 1.93 2.03 1.73 3.49 2.25 fraction; SW
Volume 46
Angiography
Cardiac Catheterization
RF
SVR
RF
0.29
1.11 0.88 1.40
0.09 0.00 0.29 0.28 0.36 0.17 0.09 0.24 0.33 0.63 0.33 0.27 0.23 0.31 0.47 0.58 0.75 0.10 0.47 0.55 0.57 0.56
0.38 0.12 0.21 0.3 1 0.24 0.30 0.20 0.25 0.64 0.28 0.49 0.18 0.32 0.52 0.66 0.68 0.48 0.51 6.42 0.71 0.56 = stroke volume ratio.
1.38 I.57 1.21 1.10 1.31 1.52 2.73 1.55 1.37 1.30 1.44 1.93 2.43 3.99 1.11 i .a9 2.21 2.32 2.27
RADIONUCLIDE ANGIOGRAPHY IN VALVE DISEASE--BOUGH ET AL.
mation of the location of volume changes within the other three cardiac chambers. With the composite functional image as a guide, the operator then redefined both ventricular regions of interest, displayed their separate time activity curves and identified the end-diastolic (ED) and end-systolic (ES) frames for each ventricle. If one assumes the ventricles to be equidistant from the camera in the left anterior oblique view and assumes a temporally constant background, the left to right ventricular stroke volume ratio (SVR) and left ventricular (LV) regurgitant fractions (RF) could be directly calculated as follow& SVR = LV (ED - ES) counts RV (ED - ES) counts RF
=SVR-l
SVR Although not strictly correct, such assumptions have proved reasonable in the clinical application of this method. In a prior series of normal subjects we found the mean (f standard deviation [SD]) value for stroke volume ratio measured in the same way to be 1.06 f 0.05, a value close to that of unity, which normal physiology demands.14J5
Results Control subjects: In the 18 control subjects without shunts or valve regurgitation, the mean stroke volume ratio was 1.00 f 0.08 (mean f SD) as calculated from cardiac catheterization data and 1.08 f 0.06 on the basis of gated radionuclide angiographic data. These values agreed closely with our previously reported results14 in 33 noncatheterized patients without clinical evidence of valve regurgitation or shunts whose average stroke volume ratio was 1.06 f 0.05 as assessed with gated radionuclide angiography. These data demonstrate that the stroke volume ratio measured in normal subjects with the gated radionuclide angiographic method closely approximates unity and that the normal range (mean f 2 SD) for this measurement is quite narrow: 0.96 to 1.20. Quantitation of regurgitant valve disease: Figure 2 shows a significant linear correlation (r = 0.79) between the stroke volume ratio determined with gated
FIGURE 1. Video display of end-diastolic frame (EDF. top) without (A) and with (B) ventricular and atrial regions of interest, and composite functional image (CFI, bottom) without XX and with (D) similar reaions of interest.
September 1980
The American Journal at CARDIOLOGY
Volume 46
425
RADIONUCLIDE ANGIOGRAPHY IN VALVE DISEASE-BOUGH ET AL
4.0
r
I I
3.0
2.5
1.0 0.5 I!
0
0.5
I
I
1
I.0
1.5
2.0
2.5
I
3.0
3.5
4.0
SVR (GRAI
radionuclide angiography and with cardiac catheterization in the 22 patients with valve regurgitation. Of the 22 patients with documented regurgitation only 1 (Case 3) had a normal value for stroke volume ratio (1.13) by gated radionuclide angiography, whereas 4 (Cases 1, 2,7 and 18) had a normal value by catheterization. Therefore, gated radionuclide angiography appeared to be slightly more sensitive than quantitative volumetric methods (95 versus 82 percent) in detecting an abnormal stroke volume ratio caused by valve regurgitation.
I.0
Comparison with semiquantitative angiography: Quantitative measurements of valve regurgitation obtained with both gated radionuclide angiography and cardiac catheterization were also compared with the results of semiquantitative aortic root and left ventricular contrast cineangiography. Because the cineangiographic grading systems depend primarily on the volume of the regurgitant jet of contrast material for interpretation, it seemed more appropriate to use the regurgitant fraction of the left ventricular stroke volume rather than the stroke volume ratio as the quantitative
-
1.0 Y =
0.6
FIGURE 2. Relation between stroke volume ratios (SVR) calculated from gated radionuclide angiography (GRA) and cardiac catheterization (CATH) data. Normal subjects are represented by circles, patients with valve regurgitation by triangles.Dashed lines indicate upper range of normal values. Correlation data calculated only for patients with valve regurgitation.
0.09x*
r = 0.72 p<.OOl SEE x0.13
-
-
(A)
0.13
06
-
Y = O.IOX l0.05 r = 0.71 p < ,001 SEE 80.14
(B)
A
?
0.6 -
z
z% :
0.6
-
0.4
-
0.2
-
s 0 :
0.4 -
0.2 P
0 0
I
I
I*
2* ANGlOGRAPHIC
I
I
3+
4*
I
5+
I
Ol
6+
0
A
I
I
I
I
I
I+
2+
3+
4+
5+
6+
ANGIOGRAPHIC
SCORE
SCORE
FIGURE 3. Relations between semiquantitative cineangiographic grading of valve regurgitation and left ventricular regurgitant fraction (RF) determined from (A) gated radionuclide angiography (GRA) and (6) cardiac catheterization (CATH) data. Control subjects are represented by circles, patients with valve regurgitation by Mangles.Correlation data are calculated only for patients with valve regurgitation.
426
September 1980
The American Journal of CARDIOLOGY
Volume 46
RADIONUCLIDE ANGIOGRAPHY IN VALVE DISEASE-BOUGH ET AL.
index. Although the regurgitant fraction is derived from exactly the same data (see Methods), it is not a linear function of the stroke volume ratio. Figure 3 shows the linear correlations between the total cineangiographic grading score and the regurgitant fraction determined with gated radionuclide angiography (r = 0.72) and with catheterization (r = 0.71) for the 22 patients with valve regurgitation. Discussion This study demonstrates that gated radionuclide angiography is an accurate noninvasive method for quantitating left ventricular volume overload from valve regurgitation and that it compares favorably with standard cardiac catheterization methods. Furthermore, because gated radionuclide angiographic measurement of the stroke volume ratio has such a narrow normal range (0.96 to 1.20), it may also be useful in detecting relatively mild valve regurgitation. In this study, 21 of 22 patients with valve regurgitation, including 3 of 4 with only l+ mitral regurgitation, had an abnormal stroke volume ratio as assessed with gated radionuclide angiography. Conversely, a normal stroke volume ratio obtained with this radionuclide method would appear to exclude hemodynamically significant aortic or mitral regurgitation. Potential limitations of method: There are several limitations to the usefulness of this technique. Left ventricular volume overload cannot be measured in the presence of right ventricular volume overload, whether the latter is due to shunt or to valve regurgitation. Another limitation is that, when both left heart valves are regurgitant, only the total left ventricular volume overload can be measured, not the relative contributions of each valve. The final limitation, common to all gated cardiac scans, is that it cannot be performed in patients with widely varying R-R intervals such as those with atria1 fibrillation or multifocal atria1 tachycardia, and
therefore it is not of use in many patients with valvular heart disease. Early radionuclide methods for quantitating valve regurgitation either required invasive catheterization for central radionuclide injection or were subject to large measurement variance.16J7 More recently, standard gated cardiac blood pool scans have been successfully used to quantitate valve regurgitation by measurement of the stroke volume ratio.8 Our study supports this latter finding and suggests some technical refinements for possible improvements in this basic radionuclide method. Clinical applications: Accurate measurement of the natural progression of aortic and mitral regurgitation and the appropriate timing of surgical interventions are difficult.la20 Recently, gated radionuclide angiography during exercise has been shown to be particularly useful for detecting early left ventricular dysfunction in patients with aortic regurgitation.21 Now, this basic radionuclide technique can be applied to the measurement of left ventricular volume overload as well. Ultimately, longitudinal studies of patients with gated radionuclide angiography, using the resting stroke volume ratio to quantitate left ventricular volume overload and using exercise-induced changes in ejection fraction to quantitate left ventricular dysfunction, should allow more accurate definition of the natural history and surgical therapy of acute and chronic regurgitant valve disease than has previously been possible. In addition to its already demonstrated usefulness in valvular heart disease, the measurement of stroke volume ratios with gated radionuclide angiography has obvious applications in congenital heart disease. Similar methods should allow simple and accurate quantitation of unidirectional shunts that are not at the ventricular level, such as those occurring in atria1 septal defect, anomalous pulmonary venous drainage or patent ductus arteriosus.22
References 1. Dexter L. Profiles in valvular disease. In: Grossman W, ed. Cardiac Catheterization and Angiography. Philadelphia: Lea 8 Febiger, 1974: 253-68. 2. Brandf PWT, O’Brien KP, Giancy DL. Cardiac catheterization: III. Angiocardiography. Au&al Radio1 1970;14:398-408. 3. Baron MG. Angiocardiographic evaluation of valvular insufficiency. Circulation 1971;43:599-805. 4. Sandler H, Dodge HT, Hay RE, Rackley CE. Quantitation of valvular insufficiency in man by angiocardiography. Am Heart J 1963;65:501-13. 5. DeMaria AN, Neumann A, Lee G, Mason DT. Mitral valve disease revisited. In: Kotlee MN, Segal EL. Clinical Echocardiography. Philadelphia: FA Davis, 1978: 59-84. 6. Abbasi AS, Allen MW, DeCristofaro D, Ungar I. Detection and quantitation of the degree of mitral regurgitation by range-gated pulsed Doppler echocardiography. Circulation 1980;61:143-7. 7. Waggoner AD, Gulnones MA, Verani MS, Miller RR. Pulsed Doppler echocardiographic detection of tricuspid insufficiency: diagnostic sensitivity and correlation with right ventricular hemodynamics (abstr). Circulation 1978;58:Suppl ll:ll-41. 8. Rigo P, Alderson PO, Robertson RM, Becker LC, Wagner HN. Measurement of aortic and mitral regurgitation by gated cardiac
blood pool scans. Circulation 1979:60:306-12. 9. Sorensen So, Granes 8, O’Rourke R, Chaudhuri 1. Non-invasive quantitation of valvular regurgitation by gated equilibrium radionuclide angiography (abstr). J Nucl Med 1979;20:626. 10. Kennedy JW. Trenholme SE, Kasser IS. Left ventricular volume and mass from single-plane cineangiocardiogram. A comparison of anteroposterior and right anterior oblique methods. Am Heart J 1970;80:343-52. 11. Cohn LH, Mason DT, Ross J, Morrow AG, Braunwald E. Preoperative assessment of aortic regurgitation in patients with mitral valve disease. Am J Cardiol 1967; 19: 177-82. 12. Sellers RD, Levy MJ, Ampiatz K, Liilehei CW. Left retrograde cardioangiography in acquired cardiac disease. Technic, indications and interpretations in 700 cases. Am J Cardiol 1969;14:43747. 13. Hegg FN, Hamilton GW, Larson SM, Ritchie JL, Richards P. Cardiac chamber imaging: a comparison of red blood cells labeled with Tcssm in vitro and in vivo. J Nucl bled 1978; 19: 129-34. 14. Bough E, Gandsman E, Winkier M, Shuiman R. Use of computer-generated functional images to define ventricular boundaries in gated radionuclide angiography. In: Computers in Cardiology. IEEE Computer Society, Long Beach, CA, 1979;3-8.
September 1980
The American Journal of CARDIOLOGY
Volume 46
427
RADIONUCLIDE
ANGIOGRAPHY
IN VALVE DISEASE-BOUGH
ET AL
15. Gandsman E, Bough E, North D, Shuiman Ft. The ventricular stroke
16.
17. 18.
19.
426
volume ratio: req&e&ents for accurate quantitation (abstr). Circulation 1979;6O:Suppl ll:ll-62. Kirch DL, Metz CE, Steele PP. Quantitation of valvular insufficiency by computerized radionuclide angiocardiography. Am J Cardiol 1974;34:711-21. Weber PM, DosRemidios LV, Josko IA. Quantitative radioisotopic angiocardiography. J Nucl Med 1972;13:815-22. Spagnuoio M, Kioth H, Taranta A, Doyle E, Pasternack 6. Natural history of rheumatic aortic regurgitation. Criteria predictive of death, congestive heart failure, and angina in young patients. Circulation 1971;44:368-0G. Kirklin JW, Pacific0 AD. Surgery for acquired valvular heart dis-
September 1980
The American Journal of CARDIOLOGY
ease. N Engl J Med 1973;288:133-40, 194-9. 20. Samueis DA, Friediich AL, Curfman GD, Buckley MJ, Austen WG. Valve replacement for aortic regurgitation: long term follow-up with factors influencing results (abstr). Circulation 1977;56:Suppl III: 111-28. 21. 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;42:351-7. 22. Bough EW, Gandsman EJ, Benham ID, Boden WE, North DL, Shuiman RS. Detection and quantitation of atrial shunts in adults by gated radionuclide angiography (abstr). Am J Cardiol 1980; 45:409.
Volume 46
EDWARD
W. BOUGH, MD, FACC
ELAN J. GANDSMAN,
PhD
DAVID L. NORTH, MS RICHARD S. SHULMAN,
MD, FACC
Providence, Rhode Island
From the Departments of Cardiology and Nuclear Medicine, The Miriam Hospital and Brown University, Providence, Rhode Island. Manuscript received January, 29, 1980; revised manuscript received April 14, 1980, accepted April 18, 1980. Address for reprints: Edward W. Bough, MD, Department of Cardiology, The Miriam Hospital, 184 Summit Avenue, Providence, Rhode Island 02906.
Gated radionuclide angiography is a new noninvasive technique that can be used to calculate the ratio of lefl and right ventricular stroke volumes. This stroke volume ratio, which must be unity in normal subjects,increases in patients with aortlc or mitral regurgitation in’direct proportion to the degree of left ventricular volume overload, provided no shunts or regurgitant right heart lesions are present. In 22 patients with aortic or mitral regurgitation there was excellent correlation between the stroke volume ratio determined with gated radlonuclide angiography and with standard quantitative catheterization methods (r = 0.79). Measurement of valve regurgitation with this radlonucllde method also correlated well with data obtained from semlquantltatlve aortic root or left ventricular clneanglography (r = 0.72). Twenty-one of the 22 patients with valve regurgitation had an abnormally elevated stroke volume ratio, thereby suggesting that gated radlonuclide angiography may be useful in detecting or excluding hemodynamically significant valve regurgitation.
Semiquantitative aortic root or left ventricular contrast cineangiography is the most common method of evaluating valve regurgitation.1-3 Combined quantitative ventriculography and cardiac output determinations allow more accurate quantitation of such regurgitation, but this technique is more complex and the results are less reproducible.4 Previously available noninvasive methods have been inadequate for accurate quantitation of valve regurgitation. Although M mode echocardiography is invaluable for demonstrating structural valve abnormalities, it allows only indirect inferences as to the severity of regurgitant lesions.5 However, pulsed Doppler echocardiography has recently been shown to be useful for localizing regurgitant lesions and for quantitating mitral regurgitation.6*7 Recent studiessyg using methods similar to ours have suggested that gated radionuclide angiography is a superior noninvasive technique for quantitating valve regurgitation. In this study we compare results of gated radionuclide angiography with those of both quantitative and semiquantitative cardiac catheterization methods in evaluating valvular heart disease in 40 patients. Methods Patients: Between November 1978 and June 1979, forty patients underwent gated radionuclide angiography within 48 hours of diagnostic right and left heart catheterization. Eighteen patients (14 male and 4 female) who were being evaluated for coronary artery disease had no angiographic or clinical evidence of valve regurgitation or shunts and served as a control group. Twenty-two pa-
September 1980
The American Journal of CARDIOLOGY
Volume 48
423
RADIONUCLIDE ANGIOGRAPHY IN VALVE DISEASE-BOUGH
ET AL
tients had typical clinical findings of aortic (n = 3), mitral (n = 10) or combined (n = 9) valve regurgitation that were confirmed at cardiac catheterization; none had clinical or catheterization evidence of tricuspid regurgitation (Table I). Cardiac catheterization and angiography: All patients underwent right and left heart catheterization in a fasting, mildly sedated state. Cardiac output and effective forward stroke volume were determined from multiple indicatordilution curves and simultaneously recorded heart rate. Quantitative single plane left ventricular cineangiograms were performed in the right anterior oblique projection and analyzed for ventricular volumes and ejection fraction in a standard mariner.... Aortic root cineangiography was performed in all patients with rheumatic valve disease and in other patients with clinical evidence of aortic regurgitation. In the absence of right heart regurgitant valve lesions anti shunts, it can be assumed that the right ventricular stroke volume (RVSV) is equal to the effective forward stroke volume (SV) determined from the cardiac output. Because the total left ventricular stroke volume (LVSV) was independently measured from quantitative cineangiographic data, the ratio of left to right ventricular stroke volumes, or stroke volume ratio (SVR), could be calculated as follows: LVSV (angiographic)
Sjq&!E= RVSV
Forward
SV (cardiac
output)
The stroke volume ratio directly quantitates total left ventricular volume overload whether one or both left heart valves are regurgitant. The regurgitant fraction (RF) of the left ventricular stroke volume has also been used as a quantitative index of valve regurgitation and was calculated from the same data as follows:* RP = LVSV - RVSV 1*x
SVR - 1
-
LVSV
-
SVR
Aortic root and left ventricular cineangiograms were graded for the severity of aortic or mitral regurgitation by two independent observers who did not know the patients’ iden-
tity. Each valve lesion was graded from l+ to 4+ according to commonly used scales; the sum for both left heart vaives was designated as the angiographic score.‘l.12 Grading disagreements between observers were resolved by simultaneous review of the films in question. Radionuclide angiography: All patients underwent gated radionuclide angiography after in vivo red cell labeling with technetium-99m pyrophosphate.‘:’ Cardiac imaging was performed with a gamma camera (Elscint LF) and a 30’ caudal slant hold collimator of intermediate sensitivity (Engineering Dynamics Corporation) in the left anterior oblique view that provided optimal ventricular separation. Gated images were collected with a computer (Digital Equipment Corporation, PDP 11/40) and an electrocardiographic synchronizer (Brattle Instruments) in a 64 by 64 matrix at a rate of at least 1S frames/s to a total of 1 million counts/frame; the total acquisition time was usually about 10 minutes. All radionuclide studies were analyzed by an operator who was unaware of the catheterization results. End-diastolic and end-systolic frames were determined from the time-activity curve of an approximated left ventricular region of interest and were used to generate two functional images by computer frame arithmetic. Subtraction of the end-diastolic from the end-systolic frame produced a ventricular stroke volume image in which each pixel displays counts proportional to ventricular volume change. Because maximal atria1 volume occurs at ventricular end-systole and minimal atria1 volume at ventricular end-diastole, “reverse” subtraction of the end-systolic from the end-diastolic frame produced an atria1 volume change image analogous to the ventricular stroke volume image.14 Computer superimposition of these two images results in a composite functional image with excellent delineation of atrioventricular borders and the ventricular regions, especially the right ventricular conus (Fig. 1). Although the composite functional image is generated from the end-diastolic and end-systolic frames of the left ventricular time-activity curve and is strictly accurate only for the left ventricular stroke volume, it provides an excellent approxi-
TABLE I Angiographic, Radionuclide and Catheterization Data in 22 Patients With Valve Regurgitation Gated Radionuclide
Angiography Lesion Grade
1+ MR l+ l-l- MR
1 1
2+ l+ MR 2+ MR 2+ MR 2+ MR 3+ MR
: 2 2
3+ 4+ MR AR 3+ AR 3+ AR 1+ MR, 2+MR, l+ MR, l+ MR, 2+ MR, l+ MR, 2+ MR, 2+ MR, 3+ MR,
:
AR = aortic regurgitation;
424
Score
September 1980
;
: 2+ AR l+AR 3+ AR 3f AR 2+ AR 4+ AR 3+ AR 3+ AR 3+ AR MR = mitral regurgitation:
3” 4 4 4 5 2 6 RF = regurgitant
The American Journal of CARDIOLOGY
SW? 1.35 1.60 1.13 1.27 1.44 1.31 1.43 1.25 1.33 2.83 i .3a 1.96 1.23 1.47 2.08 2.89 3.15 1.93 2.03 1.73 3.49 2.25 fraction; SW
Volume 46
Angiography
Cardiac Catheterization
RF
SVR
RF
0.29
1.11 0.88 1.40
0.09 0.00 0.29 0.28 0.36 0.17 0.09 0.24 0.33 0.63 0.33 0.27 0.23 0.31 0.47 0.58 0.75 0.10 0.47 0.55 0.57 0.56
0.38 0.12 0.21 0.3 1 0.24 0.30 0.20 0.25 0.64 0.28 0.49 0.18 0.32 0.52 0.66 0.68 0.48 0.51 6.42 0.71 0.56 = stroke volume ratio.
1.38 I.57 1.21 1.10 1.31 1.52 2.73 1.55 1.37 1.30 1.44 1.93 2.43 3.99 1.11 i .a9 2.21 2.32 2.27
RADIONUCLIDE ANGIOGRAPHY IN VALVE DISEASE--BOUGH ET AL.
mation of the location of volume changes within the other three cardiac chambers. With the composite functional image as a guide, the operator then redefined both ventricular regions of interest, displayed their separate time activity curves and identified the end-diastolic (ED) and end-systolic (ES) frames for each ventricle. If one assumes the ventricles to be equidistant from the camera in the left anterior oblique view and assumes a temporally constant background, the left to right ventricular stroke volume ratio (SVR) and left ventricular (LV) regurgitant fractions (RF) could be directly calculated as follow& SVR = LV (ED - ES) counts RV (ED - ES) counts RF
=SVR-l
SVR Although not strictly correct, such assumptions have proved reasonable in the clinical application of this method. In a prior series of normal subjects we found the mean (f standard deviation [SD]) value for stroke volume ratio measured in the same way to be 1.06 f 0.05, a value close to that of unity, which normal physiology demands.14J5
Results Control subjects: In the 18 control subjects without shunts or valve regurgitation, the mean stroke volume ratio was 1.00 f 0.08 (mean f SD) as calculated from cardiac catheterization data and 1.08 f 0.06 on the basis of gated radionuclide angiographic data. These values agreed closely with our previously reported results14 in 33 noncatheterized patients without clinical evidence of valve regurgitation or shunts whose average stroke volume ratio was 1.06 f 0.05 as assessed with gated radionuclide angiography. These data demonstrate that the stroke volume ratio measured in normal subjects with the gated radionuclide angiographic method closely approximates unity and that the normal range (mean f 2 SD) for this measurement is quite narrow: 0.96 to 1.20. Quantitation of regurgitant valve disease: Figure 2 shows a significant linear correlation (r = 0.79) between the stroke volume ratio determined with gated
FIGURE 1. Video display of end-diastolic frame (EDF. top) without (A) and with (B) ventricular and atrial regions of interest, and composite functional image (CFI, bottom) without XX and with (D) similar reaions of interest.
September 1980
The American Journal at CARDIOLOGY
Volume 46
425
RADIONUCLIDE ANGIOGRAPHY IN VALVE DISEASE-BOUGH ET AL
4.0
r
I I
3.0
2.5
1.0 0.5 I!
0
0.5
I
I
1
I.0
1.5
2.0
2.5
I
3.0
3.5
4.0
SVR (GRAI
radionuclide angiography and with cardiac catheterization in the 22 patients with valve regurgitation. Of the 22 patients with documented regurgitation only 1 (Case 3) had a normal value for stroke volume ratio (1.13) by gated radionuclide angiography, whereas 4 (Cases 1, 2,7 and 18) had a normal value by catheterization. Therefore, gated radionuclide angiography appeared to be slightly more sensitive than quantitative volumetric methods (95 versus 82 percent) in detecting an abnormal stroke volume ratio caused by valve regurgitation.
I.0
Comparison with semiquantitative angiography: Quantitative measurements of valve regurgitation obtained with both gated radionuclide angiography and cardiac catheterization were also compared with the results of semiquantitative aortic root and left ventricular contrast cineangiography. Because the cineangiographic grading systems depend primarily on the volume of the regurgitant jet of contrast material for interpretation, it seemed more appropriate to use the regurgitant fraction of the left ventricular stroke volume rather than the stroke volume ratio as the quantitative
-
1.0 Y =
0.6
FIGURE 2. Relation between stroke volume ratios (SVR) calculated from gated radionuclide angiography (GRA) and cardiac catheterization (CATH) data. Normal subjects are represented by circles, patients with valve regurgitation by triangles.Dashed lines indicate upper range of normal values. Correlation data calculated only for patients with valve regurgitation.
0.09x*
r = 0.72 p<.OOl SEE x0.13
-
-
(A)
0.13
06
-
Y = O.IOX l0.05 r = 0.71 p < ,001 SEE 80.14
(B)
A
?
0.6 -
z
z% :
0.6
-
0.4
-
0.2
-
s 0 :
0.4 -
0.2 P
0 0
I
I
I*
2* ANGlOGRAPHIC
I
I
3+
4*
I
5+
I
Ol
6+
0
A
I
I
I
I
I
I+
2+
3+
4+
5+
6+
ANGIOGRAPHIC
SCORE
SCORE
FIGURE 3. Relations between semiquantitative cineangiographic grading of valve regurgitation and left ventricular regurgitant fraction (RF) determined from (A) gated radionuclide angiography (GRA) and (6) cardiac catheterization (CATH) data. Control subjects are represented by circles, patients with valve regurgitation by Mangles.Correlation data are calculated only for patients with valve regurgitation.
426
September 1980
The American Journal of CARDIOLOGY
Volume 46
RADIONUCLIDE ANGIOGRAPHY IN VALVE DISEASE-BOUGH ET AL.
index. Although the regurgitant fraction is derived from exactly the same data (see Methods), it is not a linear function of the stroke volume ratio. Figure 3 shows the linear correlations between the total cineangiographic grading score and the regurgitant fraction determined with gated radionuclide angiography (r = 0.72) and with catheterization (r = 0.71) for the 22 patients with valve regurgitation. Discussion This study demonstrates that gated radionuclide angiography is an accurate noninvasive method for quantitating left ventricular volume overload from valve regurgitation and that it compares favorably with standard cardiac catheterization methods. Furthermore, because gated radionuclide angiographic measurement of the stroke volume ratio has such a narrow normal range (0.96 to 1.20), it may also be useful in detecting relatively mild valve regurgitation. In this study, 21 of 22 patients with valve regurgitation, including 3 of 4 with only l+ mitral regurgitation, had an abnormal stroke volume ratio as assessed with gated radionuclide angiography. Conversely, a normal stroke volume ratio obtained with this radionuclide method would appear to exclude hemodynamically significant aortic or mitral regurgitation. Potential limitations of method: There are several limitations to the usefulness of this technique. Left ventricular volume overload cannot be measured in the presence of right ventricular volume overload, whether the latter is due to shunt or to valve regurgitation. Another limitation is that, when both left heart valves are regurgitant, only the total left ventricular volume overload can be measured, not the relative contributions of each valve. The final limitation, common to all gated cardiac scans, is that it cannot be performed in patients with widely varying R-R intervals such as those with atria1 fibrillation or multifocal atria1 tachycardia, and
therefore it is not of use in many patients with valvular heart disease. Early radionuclide methods for quantitating valve regurgitation either required invasive catheterization for central radionuclide injection or were subject to large measurement variance.16J7 More recently, standard gated cardiac blood pool scans have been successfully used to quantitate valve regurgitation by measurement of the stroke volume ratio.8 Our study supports this latter finding and suggests some technical refinements for possible improvements in this basic radionuclide method. Clinical applications: Accurate measurement of the natural progression of aortic and mitral regurgitation and the appropriate timing of surgical interventions are difficult.la20 Recently, gated radionuclide angiography during exercise has been shown to be particularly useful for detecting early left ventricular dysfunction in patients with aortic regurgitation.21 Now, this basic radionuclide technique can be applied to the measurement of left ventricular volume overload as well. Ultimately, longitudinal studies of patients with gated radionuclide angiography, using the resting stroke volume ratio to quantitate left ventricular volume overload and using exercise-induced changes in ejection fraction to quantitate left ventricular dysfunction, should allow more accurate definition of the natural history and surgical therapy of acute and chronic regurgitant valve disease than has previously been possible. In addition to its already demonstrated usefulness in valvular heart disease, the measurement of stroke volume ratios with gated radionuclide angiography has obvious applications in congenital heart disease. Similar methods should allow simple and accurate quantitation of unidirectional shunts that are not at the ventricular level, such as those occurring in atria1 septal defect, anomalous pulmonary venous drainage or patent ductus arteriosus.22
References 1. Dexter L. Profiles in valvular disease. In: Grossman W, ed. Cardiac Catheterization and Angiography. Philadelphia: Lea 8 Febiger, 1974: 253-68. 2. Brandf PWT, O’Brien KP, Giancy DL. Cardiac catheterization: III. Angiocardiography. Au&al Radio1 1970;14:398-408. 3. Baron MG. Angiocardiographic evaluation of valvular insufficiency. Circulation 1971;43:599-805. 4. Sandler H, Dodge HT, Hay RE, Rackley CE. Quantitation of valvular insufficiency in man by angiocardiography. Am Heart J 1963;65:501-13. 5. DeMaria AN, Neumann A, Lee G, Mason DT. Mitral valve disease revisited. In: Kotlee MN, Segal EL. Clinical Echocardiography. Philadelphia: FA Davis, 1978: 59-84. 6. Abbasi AS, Allen MW, DeCristofaro D, Ungar I. Detection and quantitation of the degree of mitral regurgitation by range-gated pulsed Doppler echocardiography. Circulation 1980;61:143-7. 7. Waggoner AD, Gulnones MA, Verani MS, Miller RR. Pulsed Doppler echocardiographic detection of tricuspid insufficiency: diagnostic sensitivity and correlation with right ventricular hemodynamics (abstr). Circulation 1978;58:Suppl ll:ll-41. 8. Rigo P, Alderson PO, Robertson RM, Becker LC, Wagner HN. Measurement of aortic and mitral regurgitation by gated cardiac
blood pool scans. Circulation 1979:60:306-12. 9. Sorensen So, Granes 8, O’Rourke R, Chaudhuri 1. Non-invasive quantitation of valvular regurgitation by gated equilibrium radionuclide angiography (abstr). J Nucl Med 1979;20:626. 10. Kennedy JW. Trenholme SE, Kasser IS. Left ventricular volume and mass from single-plane cineangiocardiogram. A comparison of anteroposterior and right anterior oblique methods. Am Heart J 1970;80:343-52. 11. Cohn LH, Mason DT, Ross J, Morrow AG, Braunwald E. Preoperative assessment of aortic regurgitation in patients with mitral valve disease. Am J Cardiol 1967; 19: 177-82. 12. Sellers RD, Levy MJ, Ampiatz K, Liilehei CW. Left retrograde cardioangiography in acquired cardiac disease. Technic, indications and interpretations in 700 cases. Am J Cardiol 1969;14:43747. 13. Hegg FN, Hamilton GW, Larson SM, Ritchie JL, Richards P. Cardiac chamber imaging: a comparison of red blood cells labeled with Tcssm in vitro and in vivo. J Nucl bled 1978; 19: 129-34. 14. Bough E, Gandsman E, Winkier M, Shuiman R. Use of computer-generated functional images to define ventricular boundaries in gated radionuclide angiography. In: Computers in Cardiology. IEEE Computer Society, Long Beach, CA, 1979;3-8.
September 1980
The American Journal of CARDIOLOGY
Volume 46
427
RADIONUCLIDE
ANGIOGRAPHY
IN VALVE DISEASE-BOUGH
ET AL
15. Gandsman E, Bough E, North D, Shuiman Ft. The ventricular stroke
16.
17. 18.
19.
426
volume ratio: req&e&ents for accurate quantitation (abstr). Circulation 1979;6O:Suppl ll:ll-62. Kirch DL, Metz CE, Steele PP. Quantitation of valvular insufficiency by computerized radionuclide angiocardiography. Am J Cardiol 1974;34:711-21. Weber PM, DosRemidios LV, Josko IA. Quantitative radioisotopic angiocardiography. J Nucl Med 1972;13:815-22. Spagnuoio M, Kioth H, Taranta A, Doyle E, Pasternack 6. Natural history of rheumatic aortic regurgitation. Criteria predictive of death, congestive heart failure, and angina in young patients. Circulation 1971;44:368-0G. Kirklin JW, Pacific0 AD. Surgery for acquired valvular heart dis-
September 1980
The American Journal of CARDIOLOGY
ease. N Engl J Med 1973;288:133-40, 194-9. 20. Samueis DA, Friediich AL, Curfman GD, Buckley MJ, Austen WG. Valve replacement for aortic regurgitation: long term follow-up with factors influencing results (abstr). Circulation 1977;56:Suppl III: 111-28. 21. 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;42:351-7. 22. Bough EW, Gandsman EJ, Benham ID, Boden WE, North DL, Shuiman RS. Detection and quantitation of atrial shunts in adults by gated radionuclide angiography (abstr). Am J Cardiol 1980; 45:409.
Volume 46