- Email: [email protected]
The effects of Valsalva maneuver on global and segmental left ventricular function in presence and absence of coronary artery disease
The effects of Valsalva maneuver on global and segmental left ventricular function in presence and absence of coronary artery disease To determine the effects of the Valsalva maneuver on global and regional left ventricular function, single-plane left ventriculograms were performed in the 30-degree right anterior oblique projection in 50 patients during normal breath holding and during the late strain phase of the Valsalva maneuver. Thirty-one patients had significant coronary artery disease (greater than 70% luminal narrowing in a major coronary artery). Ventriculograms were analyzed for determination of ejection fraction, end-diastolic, and end-systolic volumes. Regional wall motion was analyzed by a chord method of calculating segmental fractional shortening. Ejection fraction increased significantly in the entire group of patients (62 -t 16% to 70 f lg%, p < O.OOOl), while both end-diastolic (105 + 33 cc to 66 i 34 cc, p < 0.0001) and end-systolic volumes (43 f 29 cc to 30 + 29 cc, p < 0.0001) showed striking reductions with Valsalva maneuver. Patients without significant coronary disease usually exhibited global augmentation in left ventricular function, while those with coronary disease often exhibited only segmental improvement. This augmentation appeared to be dependent on the patency of the supplying coronary vessel. (AM HEART J lOg:25g, 1985.)
Arthur J. Labovitz, M.D., Bulent Dincer, M.D., Gerald Mudd, Umit T. Aker, M.D., and Harold L. Kennedy, M.D. St. Louis, MO.
The Valsalva maneuver is known to produce striking alterations in intrathoracic pressure and subsequent venous return and stroke volume.1*2 However, the effects of these changes on global and regional left ventricular function in the presence of coronary artery disease have not been well described. The various phases of Valsalva maneuver include phase I or initiation of strain, resulting in a transient rise in arterial blood pressure and a rapid decline in venous return; phase II, representing the entire duration of straining and associated with falling blood pressure and increasing heart rate; phase III or the release of strain and resultant increase in venous return; and phase IV or “overshoot” of arterial blood pressure. An observation was made in the cardiac catheterization laboratory that some patients inadvertently perform a Valsalva maneuver when asked to hold their breath for filming of the left ventriculogram. In
From cine. Received accepted Reprint University
the
Division
of Cardiology,
for publication July 10, 1984.
Feb.
St. Louis 16, 1984;
requests: Arthur Labovitz, Hospital, 1325 S. Grand
University
revision
received
School June
M.D., Division of Cardiology, Blvd., St. Louis, MO 63104.
of Medi‘7, 1984; St. Louis
M.D.,
addition, it was noted that this maneuver produced some augmentation of left ventricular systolic function. The present study was undertaken to evaluate the effect of the Valsalva maneuver on global and regional left ventricular function in the presence or absence of coronary artery disease. METHODS Patients. The study population consistedof 50 patients referred for diagnostic left heart catheterization for symptoms of chest pain (Table I). Patients with valvular or congenital heart disease,unstable angina, or overt congestive heart failure were excluded from the study. Valsalva maneuver. Valsalva maneuver was accomplished by having patients exhale into a mouthpiece attached to a manometer. Patients unstable to maintain an intra-oral pressureof at least 40 mm Hg for at least 10 secondsin a practice sessionprior to catheterization were also excluded from the study. Left ventriculography. Informed consent was obtained from all patients prior to catheterization. Single-planeleft ventriculograms were filmed in standard 30-degreeright anterior oblique projection at 60 framesfsecusing 30 to 45 cc of diatrizoate meglumineduring normal breath holding, as well as during the late strain phase of the Valsalva maneuver. The repeat ventriculogram followed the first ventriculogram by 3 to 15 minutes. In seven patients the
259
February,
260
Labovitz
et al.
American
1985
Heart Journal
R, are supplied by the left anterior descendingartery and segmentsRq,R,, and R, are supplied by the right coronary artery unless the system is left-dominant. Statistical analysiswasaccomplishedby useof dependent Student’s t test and chi square methods. There were no untoward complications of these angiographic procedures. RESULTS OL
I STRAIN
I
RELEASE
1. Characteristic arterial pressuretracing obtained during Valsalva maneuver with arrows indicating initiation and releaseof strain as well as the timing of the left ventriculogram. Fig.
Table
1. Clinical characteristics CAD
‘(# Male
No CAD
BO’,’
Age
Meanyears Range &blockers
LVEDP (mean)* HR (mean)* CAD = coronary lar end-diastolic ‘Pre angiogram.
(n = 31)
50.7 34-67 16’, 9.1 12.1
artery disease; HR = heart rate; LVEDP pressure.
(n = 19)
63°C
41.0 27-64 16’,
1.8 X1.8 = left ventricu-
Valsalva maneuver ventriculogram preceded the normal breath-holding study. Arterial blood pressurewasrecorded simultaneously from an arterial sheath positioned in the femoral artery. Left ventricular injection during the Valsalva maneuver wastimed to coincide with the hypotensive late strain phaseof the Valsalva maneuver (Fig. 1) or 10 seconds after the initiation of Valsalva in the absenceof a sigmoidal curve. The normal breath-holding ventriculogram was performed at end inspiration. Global LV function. End-diastolic and end-systolic frames were traced from 35 mm tine film, taking care not to include either ventricular extrasystolic beats or postextrasystolic beats.Left ventricular volumeswere calculated using a modification of Dodge’s area-length method.“’ Left ventricular ejection fraction was calculated for both the control (normal breath holding) ventriculogram as well as for the Valsalva maneuver ventriculogram. Segmental LV function. Regional wall motion was analyzed by a floating axis chordal method. The aortic root and apex are identified and a long-axis dimension is determined from the center of the aortic root to the apex, This long axis is then quadrasected by three equally spacedperpendicular chords, creating six radii (R, to R,). The systolic and diastolic long axesare then superimposed using a floating analysis, and chordal shortening fraction is calculated by dividing the difference between the systolic and diastolic radius by the diastolic radius (Fig. 2). For the purposeof this discussion,segmentsR,, RS,and
Cardiac disease. There were 31 patients with coronary artery disease (greater than 70% luminal narrowing in a major coronary vessel) and 19 patients without significant coronary artery lesions. There were 24 patients with significant narrowing in the right coronary artery, 27 patients with left anterior descending lesion, and 19 patients with disease in the circumflex system. Eight patients had singlevessel disease, nine had double-vessel disease, and 14 had triple-vessel disease. Of the patients without coronary artery disease, there were two with a clinical diagnosis of congestive cardiomyopathy, one with biopsy-proven myocarditis, and one with echocardiographic features of hypertrophic cardiomyopathy. The remaining 15 patients had no evidence of cardiovascular disease. Left ventricular ejection fraction. The ejection fractions of the entire group of patients obtained from the Valsalva maneuver left ventriculograms were almost uniformly increased when compared to control ventriculograms obtained during normal breath holding (69.6 t 18% versus 61.8 + IS%, p < 0.0001). There was a significant difference between control ventriculograms and Valsalva maneuver ventriculograms in patients with (57 i15 % to 63.3 t- 18%) p < 0.0001) and without coronary artery disease (69.5 4 16% to 79.8 4 15%) p < 0.0002). This augmentation with Valsalva maneuver (11% in patients with coronary artery disease and 15 % in patients without coronary artery disease) was similar in both groups (Fig. 3). As might be expected, the patients without coronary artery disease had greater ejection fractions than those patients with coronary artery disease. Left ventricular volumes. Left ventricular enddiastolic and end-systolic volumes were significantly reduced in most patients with Valsalva maneuver. In the entire group of patients, there was a decrease in end-diastolic volume from 104.5 +- 32 cc to 87.9 f 34 cc (p < 0.0001). A similar significant decrease in end-diastolic volume with Valsalva maneuver was found in patients without coronary artery disease (104.8 + 36 cc to 89 f 29 cc, p < 0.0005) and with coronary artery disease (104.3 I 31 cc to 87.3 -+ 37 cc, p < 0.0001). Changes in end-systolic volume were even more striking, with an overall decrease from 42.7 k 29 cc to 30.0 -t 29 cc (p < 0.0001) in the
Volume Number
109 2
Valsalva and LV function
in CAD
261
Anterior
Posterior SF1
: IRD,
- Rs, I/RD,
SF2
i 1~~2 SF3 j IRD~
- RS21/~~2 - RS31f~~3
SF4 SF5
: lRD4
. RS41/RD4
i
IRDS
- RS6l/RD6
sF6
:
lRD6
- RS61/RD6
Fig. 2. Chord method for analyzing segmental wall motion. There are six zonal radii (I&, F&,. . .). RD = diastolic radius; RS = systolic radius: SF = shortening fraction.
EJECTIO~~~RACTION NO CAD
20
1 I CONTROL
I I “ALSALVI
END DIASTOLIC
VOLUME
CAD “41
I I CONTROL
I I “ALSAWA
Fig. 3. Left ventricular ejection fractions calculated from ventriculograms obtained during normal breath holding (CONTROL) and during the late strain phaseof Valsalva maneuver.
in end-systolic volume resulting from Valsalva maneuver in patients with (46.7 -C 27 cc to 35.1 & 30 cc, p < 0.0001) and without coronary artery disease (36.3 f 32 cc to 21.8 f 27 cc, p < 0.0001) were similar (Fig. 4). Segmental wall motion. The mean fractional shortening for the six segments analyzed in patients without detectable heart disease is shown in Table II and was used as a baseline standard in these observations. In the entire group of patients, 191(64%) of the 300 segments analyzed showed increased fractional shortening with Valsalva maneuver. Patients without coronary artery disease typically exhibited a global augmentation of left ventricular function,
END SYSTOLIC VOLUME lmll ‘P
NO CAD ml9
T
entire group of patients. Changes
CONTRclL
II
“ALSAWA
Fig. 4. Left ventricular end-diastolic volume (‘A) and
end-systolic volume (B) during normal breath holding (CONTROL) and during the late strain phase of the Valsalva maneuver.
February.
262
Labovitz et al.
American
1985
Heart
Journal
“ALSALW
CONTROL NORMAL
CORONARY ARTERIES
EF:
EF:081
065
VALSALVA
CONTROL
VALSALVA
SDXRCA
70%
EF=
lOD%RCA gO%LAD
LAD
EF=
EF-084
0.53
04,
EF=O64
Fig. 5. Digitized reproduction of ventriculograms performed during normal breath holding
(left) and during the left strain phaseof the Valsalva maneuver (right). A, Normal coronary arteries. Note global augmentation of systolic function. B, Subtotal occlusion of right coronary artery and left anterior descendingcoronary artery. Note improved contraction during Valsalva maneuver in regionssupplied by both vessels.C, Total occlusion of right coronary artery and subtotal occlusion of the left anterior descendingcoronary artery. Only the segmentssupplied by left anterior descendingcoronary artery show normal systolic function during Valsalva maneuver.
Table
II. Segmentalshortening in patients without heart disease
..___
__-.~~
Segmer7t ____R,
Mean ‘,, shortening Standarddeviation
2
49.3 20.4
R2
44.6
-+ 22.3
while patients with significant coronary artery disease exhibited various degrees of segmental improvement (Fig. 5). An abnormal segment was defined as a myocardial segment with a fractional shortening of less than 25 % . There were 82 abnormal myocardial segments in patients with coronary artery disease (Fig. 6), of which 32 (39 % ) normalized to a fractional shorteliing of greater than 25%) with an increase of 30% over normal breath holding during the Valsalva maneuver left ventriculogram. Analyzing specific segments in relation to the coronary artery supplying those segments showed that segments supplied by a totally occluded coronary
RI
&
42.8
i 22.2
_.-
46.7 * 20.4
H\
43.2 i 1.5.0 -_-. .~ -
&
31.5 2 9.2
artery (33 segments) improved less frequently than those with subtotal occlusive lesions (49 segments) (Fig. 7). In fact, only 21% of the hypokinetic segments supplied by a totally occluded coronary artery normalized with Valsalva maneuver, while approximately half of the hypokinetic segments supplied by coronary arteries with less than total occlusion normalized with Valsalva maneuver. Of the 24 patients with coronary artery disease in whom hypokinetic segments were seen, 12 had ECGs consistent with transmural myocardial infarction. Four of these 12 patients (33 % ) had improvement of hypokinetic segments with Valsalva maneuver. Of
Volume
109
Number
2
Valsalva
the 12 patients with resting segmental abnormalities and no ECG evidence of myocardial infarction, eight (67 9%) had segmental improvement with Valsalva maneuver. These differences did not achieve statistical significance. Arterial
pressure
response
to Valsalva
and LV function
in CAD
263
82 HYPOKINETIC
SEGMENTS
maneuver.
There were only five patients in the study population who failed to exhibit a sigmoidal response of arterial pressure to Valsalva maneuver. These patients exhibited the so-called square wave response consisting of a transient rise in blood pressure associated with straining with a return to pre-Valsalva blood pressure with the release of strain. Two of these patients had a diagnosis of cardiomyopathy; the other three had triple-vessel coronary artery disease. All five had left ventricular ejection fractions of less than 40% at rest. Although none of these patients exhibited a significant increase in left ventricular ejection fraction with Valsalva maneuver, all but one had significant decreases in left ventricular volumes.
50
32 NORMALIZED
d TOTALLY OCCLUDED
SUB TOTAL OCCLUSION
TOTALLY OCCLUDED
Fig. 6. Diagrammatic representation of abnormal myo-
cardial segmentsand their relationship to patency of supplying coronary artery in patients with coronary artery disease.
DISCUSSION Segmental asynergy. It has been well shown that segmental wall motion abnormalities in the presence of coronary artery disease need not be secondary to myocardial infarction with resultant fibrous s~ar.~-* In fact, segmental asynergy is commonly found in regions hypoperfused by stenotic coronary arteries, and thus provides a marker of resting ischemia or jeopardized myocardium. Several methods including post extrasystolic potentiation: catecholemine infusion,‘O and nitroglycerin l1 have been employed to help differentiate potentially functional myocardial segments from truly akinetic, fibrotic segments and therefore predict the usefulness of various interventions including coronary artery bypass surgery on left ventricular function. LV contractile reserve. The present study demonstrates that ventriculography during Valsalva maneuver provides a safe and simple means of assessing segmental myocardial contractile reserve. The dramatic decrease in left ventricular end-systolic and end-diastolic volumes seen during Valsalva maneuver in this study is similar to that reported by other investigators in both angiographic and echocardiographic studies .l, 2,12,l3 The increase in global left ventricular function as measured by left ventricular ejection fraction was equally striking. We also confirmed earlier observations that a “square wave” response to Valsalva maneuver reflects severe underlying left ventricular dysfunction, since this response characterized the bulk of the patients not
7
\
SUB TOTAL OCCLUSION
m
Total
m
Sub Total
Occlusion Occlusion
6%
n-33
Iv49
Fig. 7. Percent of dysfunctional segmentsseenon control (normal breath holding) ventriculogram that normalized on Valsalva maneuver ventriculogram in the distribution of totally occluded or less than totally occluded coronary arteries.
exhibiting augmentation in left ventricular ejection fraction with Valsalva maneuver.14 Although the limitations of a chordal system for evaluating segmental wall motion quantitatively have been described,15v16 the results should be at least qualitatively reproducible, since the same system was used
February.
264
Labovitz et al.
for analyzing both normal breath holding and Valsalva maneuver left ventriculograms. Mechanisms. While it was not the purpose of this study to determine the mechanisms by which the Valsalva maneuver augments global and regional left ventricular function, we believe that there are at least two major factors involved. First, the decreased stroke volume resulting from the decreased venous return in late strain phase of the Valsalva maneuver results in arterial hypotension. This in turn stimulates arterial b’aroreceptors to enhance alpha and beta sympathetic nerve stimulation with subsequent enhanced myocardial contractility,” similar to catecholamine infusion. In addition, to explain improvement in ischemic segments, we note that there is probably a reduction in factors determining myocardial oxygen demand. Most notably, we see that during continued straining there are simultaneous reductions in both arterial blood pressure as well as in left ventricular dimensions. Thus both pressure and volume determinants of myocardial wall tension and related oxygen demand are reduced. This mechanism has been felt to be responsible for reports of relief of angina by the Valsalva maneuver.18 We therefore can hypothesize that the first mechanism, increased adrenergic discharge, is responsible for the augmentation in contraction of those myocardial segments supplied by normal coronary arteries, whereas the favorable change in myocardial oxygen demand is the predominant mechanism for enhanced contractility of myocardial segments with reduced coronary blood flow but reversible left ventricular dysfunction. Conclusions. On the basis of the present study, it would appear that the left ventricular cineangiographic changes resulting from the Valsalva maneuver may provide a predictive indicator of which asynergistic myocardial segments will improve by secondary interventions such as coronary revascularization from coronary bypass graft surgery or percutaneous transluminal angioplasty. Certainly, follow-up of such persons is clearly of current interest. In addition, it should be noted that because some patients inadvertently perform a Valsalva maneuver when told to hold their breath for left ventriculography, segmental wall motion abnormalities may be masked. The angiographer, therefore, should be aware of this effect and interpret his results accordingly.
American
We would like to thank Gina Romano Useted for secretarial assistance, and Kren analysis.
1985
Heart Journal
for artwork. J,aurelle Shriver for statistical
REFERENCES
1. Brooker .I%, Alderman EL, Harrison DC: Alterations in left ventricular volumes induced by Valsalva maneuver. Br Heart J 36:713. 1974. 2. (irose R. Masking C, Spindola-Franc0 H, Yipintsoi T: Production of left ventricular cavity obliteration in normal man. Circulation 64:448, 1981. 3. Dodge HT, Sandler H, Ballew DW, Lord JD dr: The use 1)1 biplane angiocardiography for measurement of left ventricular volume-in man. AM HEART J 60:762, 1960. 4. Greene DG. Carlisle R. Grant C. Bunneil IL: Estimation of left ventricular volume~by one-plane cineangiography. Circulation 351. 1967. 51 Rackley CE: Quantitative evaluation of left ventricular func tion by radiographic techniques. Circulation 54:862, 1976. 6. Bodenheimer MM, Banka VS, Hermann GA: Reversible asynergy: Histopathologic and electrographic correlations in patients with coronary artery disease. Circulation 53:792, 1976. 7. Ideker RE, Behar VS, Wagner GS. Starr ,JW. Starmer CF. Lee KL. Hackel DB: Evaluation of asvnerev as an indicator of myocardial fibrosis. Circulation 57:715. 1977. 8. Massie BM. Botvinick EH, Brundage BH, Greenberg B, Shames D, Gelberg H: Relationship of regional myocardial perfusion to segmental wall motion. Circulation 58:1154, 1978. 9. Dyke SH, Cohn PF, Gorlin R, Sonnenblick ED: Detection of’ residual myocardial function in coronary artery disease using post-extra systolic potentiation. Circulation 50:694, 1974. LE. Cohn PF. Herman MV. Gorlin R: 10. Horn HR. Teichholz Augmentation of left ventricular contraction pat.tern in corenary artery disease by an inotropic catecholamine. Circulation 49:1063, 1974. 11. Helfant RH. Pine R, Meister SG, Feldman MS, Trout KG. Banka VS: Nitroglycerin to unmask reversible asynergy. Circulation 50:108, 1974. 12. Parisi AF, Harrington JJ, Askenazi J, Pratt RC. McIntyre KM: Echocardiographic evaluation of the Valsalva maneuver in healthy subjects and patients with and without heart failure. Circulation 54:921, 1976. 13. Robertson D, Stevens RM, Friesinger (;c’, Gates JA: The effect of the Valsalva maneuver on echocardiographic dimensions in man. Circulation 55596. 1977. 14. Zema Md. Restivo B. SOS T. Sniderman KW. Kline S: Left ventricular dysfunction-bedside Valsalva manoeuver. Br Heart J 44~560, 1980. 1.5. Leighton RF, Wilt SM, Lewis RP: Detection of hypokinesis by a quantitative analysis of left ventricular cineangiograms. Circulation 50:121, 1974. 16. Gelberg Hd, Brundage BH, Glanz S, Parmley WW: Quantitative left ventricular wall motion analysis: A comparison of area, chord and radial methods. Circulation 59:991. 1979. sinus reflex control of coronary blood flow. 17. Feigh EO: Carotid Circ Res 23:223. 1968. 18. Pepine CJ, Wiener I,: Effects of the Valsalva maneuver on myocardial ischemia in patients with coronary artery disease. Circulation 59:1304, 1979.
Arthur J. Labovitz, M.D., Bulent Dincer, M.D., Gerald Mudd, Umit T. Aker, M.D., and Harold L. Kennedy, M.D. St. Louis, MO.
The Valsalva maneuver is known to produce striking alterations in intrathoracic pressure and subsequent venous return and stroke volume.1*2 However, the effects of these changes on global and regional left ventricular function in the presence of coronary artery disease have not been well described. The various phases of Valsalva maneuver include phase I or initiation of strain, resulting in a transient rise in arterial blood pressure and a rapid decline in venous return; phase II, representing the entire duration of straining and associated with falling blood pressure and increasing heart rate; phase III or the release of strain and resultant increase in venous return; and phase IV or “overshoot” of arterial blood pressure. An observation was made in the cardiac catheterization laboratory that some patients inadvertently perform a Valsalva maneuver when asked to hold their breath for filming of the left ventriculogram. In
From cine. Received accepted Reprint University
the
Division
of Cardiology,
for publication July 10, 1984.
Feb.
St. Louis 16, 1984;
requests: Arthur Labovitz, Hospital, 1325 S. Grand
University
revision
received
School June
M.D., Division of Cardiology, Blvd., St. Louis, MO 63104.
of Medi‘7, 1984; St. Louis
M.D.,
addition, it was noted that this maneuver produced some augmentation of left ventricular systolic function. The present study was undertaken to evaluate the effect of the Valsalva maneuver on global and regional left ventricular function in the presence or absence of coronary artery disease. METHODS Patients. The study population consistedof 50 patients referred for diagnostic left heart catheterization for symptoms of chest pain (Table I). Patients with valvular or congenital heart disease,unstable angina, or overt congestive heart failure were excluded from the study. Valsalva maneuver. Valsalva maneuver was accomplished by having patients exhale into a mouthpiece attached to a manometer. Patients unstable to maintain an intra-oral pressureof at least 40 mm Hg for at least 10 secondsin a practice sessionprior to catheterization were also excluded from the study. Left ventriculography. Informed consent was obtained from all patients prior to catheterization. Single-planeleft ventriculograms were filmed in standard 30-degreeright anterior oblique projection at 60 framesfsecusing 30 to 45 cc of diatrizoate meglumineduring normal breath holding, as well as during the late strain phase of the Valsalva maneuver. The repeat ventriculogram followed the first ventriculogram by 3 to 15 minutes. In seven patients the
259
February,
260
Labovitz
et al.
American
1985
Heart Journal
R, are supplied by the left anterior descendingartery and segmentsRq,R,, and R, are supplied by the right coronary artery unless the system is left-dominant. Statistical analysiswasaccomplishedby useof dependent Student’s t test and chi square methods. There were no untoward complications of these angiographic procedures. RESULTS OL
I STRAIN
I
RELEASE
1. Characteristic arterial pressuretracing obtained during Valsalva maneuver with arrows indicating initiation and releaseof strain as well as the timing of the left ventriculogram. Fig.
Table
1. Clinical characteristics CAD
‘(# Male
No CAD
BO’,’
Age
Meanyears Range &blockers
LVEDP (mean)* HR (mean)* CAD = coronary lar end-diastolic ‘Pre angiogram.
(n = 31)
50.7 34-67 16’, 9.1 12.1
artery disease; HR = heart rate; LVEDP pressure.
(n = 19)
63°C
41.0 27-64 16’,
1.8 X1.8 = left ventricu-
Valsalva maneuver ventriculogram preceded the normal breath-holding study. Arterial blood pressurewasrecorded simultaneously from an arterial sheath positioned in the femoral artery. Left ventricular injection during the Valsalva maneuver wastimed to coincide with the hypotensive late strain phaseof the Valsalva maneuver (Fig. 1) or 10 seconds after the initiation of Valsalva in the absenceof a sigmoidal curve. The normal breath-holding ventriculogram was performed at end inspiration. Global LV function. End-diastolic and end-systolic frames were traced from 35 mm tine film, taking care not to include either ventricular extrasystolic beats or postextrasystolic beats.Left ventricular volumeswere calculated using a modification of Dodge’s area-length method.“’ Left ventricular ejection fraction was calculated for both the control (normal breath holding) ventriculogram as well as for the Valsalva maneuver ventriculogram. Segmental LV function. Regional wall motion was analyzed by a floating axis chordal method. The aortic root and apex are identified and a long-axis dimension is determined from the center of the aortic root to the apex, This long axis is then quadrasected by three equally spacedperpendicular chords, creating six radii (R, to R,). The systolic and diastolic long axesare then superimposed using a floating analysis, and chordal shortening fraction is calculated by dividing the difference between the systolic and diastolic radius by the diastolic radius (Fig. 2). For the purposeof this discussion,segmentsR,, RS,and
Cardiac disease. There were 31 patients with coronary artery disease (greater than 70% luminal narrowing in a major coronary vessel) and 19 patients without significant coronary artery lesions. There were 24 patients with significant narrowing in the right coronary artery, 27 patients with left anterior descending lesion, and 19 patients with disease in the circumflex system. Eight patients had singlevessel disease, nine had double-vessel disease, and 14 had triple-vessel disease. Of the patients without coronary artery disease, there were two with a clinical diagnosis of congestive cardiomyopathy, one with biopsy-proven myocarditis, and one with echocardiographic features of hypertrophic cardiomyopathy. The remaining 15 patients had no evidence of cardiovascular disease. Left ventricular ejection fraction. The ejection fractions of the entire group of patients obtained from the Valsalva maneuver left ventriculograms were almost uniformly increased when compared to control ventriculograms obtained during normal breath holding (69.6 t 18% versus 61.8 + IS%, p < 0.0001). There was a significant difference between control ventriculograms and Valsalva maneuver ventriculograms in patients with (57 i15 % to 63.3 t- 18%) p < 0.0001) and without coronary artery disease (69.5 4 16% to 79.8 4 15%) p < 0.0002). This augmentation with Valsalva maneuver (11% in patients with coronary artery disease and 15 % in patients without coronary artery disease) was similar in both groups (Fig. 3). As might be expected, the patients without coronary artery disease had greater ejection fractions than those patients with coronary artery disease. Left ventricular volumes. Left ventricular enddiastolic and end-systolic volumes were significantly reduced in most patients with Valsalva maneuver. In the entire group of patients, there was a decrease in end-diastolic volume from 104.5 +- 32 cc to 87.9 f 34 cc (p < 0.0001). A similar significant decrease in end-diastolic volume with Valsalva maneuver was found in patients without coronary artery disease (104.8 + 36 cc to 89 f 29 cc, p < 0.0005) and with coronary artery disease (104.3 I 31 cc to 87.3 -+ 37 cc, p < 0.0001). Changes in end-systolic volume were even more striking, with an overall decrease from 42.7 k 29 cc to 30.0 -t 29 cc (p < 0.0001) in the
Volume Number
109 2
Valsalva and LV function
in CAD
261
Anterior
Posterior SF1
: IRD,
- Rs, I/RD,
SF2
i 1~~2 SF3 j IRD~
- RS21/~~2 - RS31f~~3
SF4 SF5
: lRD4
. RS41/RD4
i
IRDS
- RS6l/RD6
sF6
:
lRD6
- RS61/RD6
Fig. 2. Chord method for analyzing segmental wall motion. There are six zonal radii (I&, F&,. . .). RD = diastolic radius; RS = systolic radius: SF = shortening fraction.
EJECTIO~~~RACTION NO CAD
20
1 I CONTROL
I I “ALSALVI
END DIASTOLIC
VOLUME
CAD “41
I I CONTROL
I I “ALSAWA
Fig. 3. Left ventricular ejection fractions calculated from ventriculograms obtained during normal breath holding (CONTROL) and during the late strain phaseof Valsalva maneuver.
in end-systolic volume resulting from Valsalva maneuver in patients with (46.7 -C 27 cc to 35.1 & 30 cc, p < 0.0001) and without coronary artery disease (36.3 f 32 cc to 21.8 f 27 cc, p < 0.0001) were similar (Fig. 4). Segmental wall motion. The mean fractional shortening for the six segments analyzed in patients without detectable heart disease is shown in Table II and was used as a baseline standard in these observations. In the entire group of patients, 191(64%) of the 300 segments analyzed showed increased fractional shortening with Valsalva maneuver. Patients without coronary artery disease typically exhibited a global augmentation of left ventricular function,
END SYSTOLIC VOLUME lmll ‘P
NO CAD ml9
T
entire group of patients. Changes
CONTRclL
II
“ALSAWA
Fig. 4. Left ventricular end-diastolic volume (‘A) and
end-systolic volume (B) during normal breath holding (CONTROL) and during the late strain phase of the Valsalva maneuver.
February.
262
Labovitz et al.
American
1985
Heart
Journal
“ALSALW
CONTROL NORMAL
CORONARY ARTERIES
EF:
EF:081
065
VALSALVA
CONTROL
VALSALVA
SDXRCA
70%
EF=
lOD%RCA gO%LAD
LAD
EF=
EF-084
0.53
04,
EF=O64
Fig. 5. Digitized reproduction of ventriculograms performed during normal breath holding
(left) and during the left strain phaseof the Valsalva maneuver (right). A, Normal coronary arteries. Note global augmentation of systolic function. B, Subtotal occlusion of right coronary artery and left anterior descendingcoronary artery. Note improved contraction during Valsalva maneuver in regionssupplied by both vessels.C, Total occlusion of right coronary artery and subtotal occlusion of the left anterior descendingcoronary artery. Only the segmentssupplied by left anterior descendingcoronary artery show normal systolic function during Valsalva maneuver.
Table
II. Segmentalshortening in patients without heart disease
..___
__-.~~
Segmer7t ____R,
Mean ‘,, shortening Standarddeviation
2
49.3 20.4
R2
44.6
-+ 22.3
while patients with significant coronary artery disease exhibited various degrees of segmental improvement (Fig. 5). An abnormal segment was defined as a myocardial segment with a fractional shortening of less than 25 % . There were 82 abnormal myocardial segments in patients with coronary artery disease (Fig. 6), of which 32 (39 % ) normalized to a fractional shorteliing of greater than 25%) with an increase of 30% over normal breath holding during the Valsalva maneuver left ventriculogram. Analyzing specific segments in relation to the coronary artery supplying those segments showed that segments supplied by a totally occluded coronary
RI
&
42.8
i 22.2
_.-
46.7 * 20.4
H\
43.2 i 1.5.0 -_-. .~ -
&
31.5 2 9.2
artery (33 segments) improved less frequently than those with subtotal occlusive lesions (49 segments) (Fig. 7). In fact, only 21% of the hypokinetic segments supplied by a totally occluded coronary artery normalized with Valsalva maneuver, while approximately half of the hypokinetic segments supplied by coronary arteries with less than total occlusion normalized with Valsalva maneuver. Of the 24 patients with coronary artery disease in whom hypokinetic segments were seen, 12 had ECGs consistent with transmural myocardial infarction. Four of these 12 patients (33 % ) had improvement of hypokinetic segments with Valsalva maneuver. Of
Volume
109
Number
2
Valsalva
the 12 patients with resting segmental abnormalities and no ECG evidence of myocardial infarction, eight (67 9%) had segmental improvement with Valsalva maneuver. These differences did not achieve statistical significance. Arterial
pressure
response
to Valsalva
and LV function
in CAD
263
82 HYPOKINETIC
SEGMENTS
maneuver.
There were only five patients in the study population who failed to exhibit a sigmoidal response of arterial pressure to Valsalva maneuver. These patients exhibited the so-called square wave response consisting of a transient rise in blood pressure associated with straining with a return to pre-Valsalva blood pressure with the release of strain. Two of these patients had a diagnosis of cardiomyopathy; the other three had triple-vessel coronary artery disease. All five had left ventricular ejection fractions of less than 40% at rest. Although none of these patients exhibited a significant increase in left ventricular ejection fraction with Valsalva maneuver, all but one had significant decreases in left ventricular volumes.
50
32 NORMALIZED
d TOTALLY OCCLUDED
SUB TOTAL OCCLUSION
TOTALLY OCCLUDED
Fig. 6. Diagrammatic representation of abnormal myo-
cardial segmentsand their relationship to patency of supplying coronary artery in patients with coronary artery disease.
DISCUSSION Segmental asynergy. It has been well shown that segmental wall motion abnormalities in the presence of coronary artery disease need not be secondary to myocardial infarction with resultant fibrous s~ar.~-* In fact, segmental asynergy is commonly found in regions hypoperfused by stenotic coronary arteries, and thus provides a marker of resting ischemia or jeopardized myocardium. Several methods including post extrasystolic potentiation: catecholemine infusion,‘O and nitroglycerin l1 have been employed to help differentiate potentially functional myocardial segments from truly akinetic, fibrotic segments and therefore predict the usefulness of various interventions including coronary artery bypass surgery on left ventricular function. LV contractile reserve. The present study demonstrates that ventriculography during Valsalva maneuver provides a safe and simple means of assessing segmental myocardial contractile reserve. The dramatic decrease in left ventricular end-systolic and end-diastolic volumes seen during Valsalva maneuver in this study is similar to that reported by other investigators in both angiographic and echocardiographic studies .l, 2,12,l3 The increase in global left ventricular function as measured by left ventricular ejection fraction was equally striking. We also confirmed earlier observations that a “square wave” response to Valsalva maneuver reflects severe underlying left ventricular dysfunction, since this response characterized the bulk of the patients not
7
\
SUB TOTAL OCCLUSION
m
Total
m
Sub Total
Occlusion Occlusion
6%
n-33
Iv49
Fig. 7. Percent of dysfunctional segmentsseenon control (normal breath holding) ventriculogram that normalized on Valsalva maneuver ventriculogram in the distribution of totally occluded or less than totally occluded coronary arteries.
exhibiting augmentation in left ventricular ejection fraction with Valsalva maneuver.14 Although the limitations of a chordal system for evaluating segmental wall motion quantitatively have been described,15v16 the results should be at least qualitatively reproducible, since the same system was used
February.
264
Labovitz et al.
for analyzing both normal breath holding and Valsalva maneuver left ventriculograms. Mechanisms. While it was not the purpose of this study to determine the mechanisms by which the Valsalva maneuver augments global and regional left ventricular function, we believe that there are at least two major factors involved. First, the decreased stroke volume resulting from the decreased venous return in late strain phase of the Valsalva maneuver results in arterial hypotension. This in turn stimulates arterial b’aroreceptors to enhance alpha and beta sympathetic nerve stimulation with subsequent enhanced myocardial contractility,” similar to catecholamine infusion. In addition, to explain improvement in ischemic segments, we note that there is probably a reduction in factors determining myocardial oxygen demand. Most notably, we see that during continued straining there are simultaneous reductions in both arterial blood pressure as well as in left ventricular dimensions. Thus both pressure and volume determinants of myocardial wall tension and related oxygen demand are reduced. This mechanism has been felt to be responsible for reports of relief of angina by the Valsalva maneuver.18 We therefore can hypothesize that the first mechanism, increased adrenergic discharge, is responsible for the augmentation in contraction of those myocardial segments supplied by normal coronary arteries, whereas the favorable change in myocardial oxygen demand is the predominant mechanism for enhanced contractility of myocardial segments with reduced coronary blood flow but reversible left ventricular dysfunction. Conclusions. On the basis of the present study, it would appear that the left ventricular cineangiographic changes resulting from the Valsalva maneuver may provide a predictive indicator of which asynergistic myocardial segments will improve by secondary interventions such as coronary revascularization from coronary bypass graft surgery or percutaneous transluminal angioplasty. Certainly, follow-up of such persons is clearly of current interest. In addition, it should be noted that because some patients inadvertently perform a Valsalva maneuver when told to hold their breath for left ventriculography, segmental wall motion abnormalities may be masked. The angiographer, therefore, should be aware of this effect and interpret his results accordingly.
American
We would like to thank Gina Romano Useted for secretarial assistance, and Kren analysis.
1985
Heart Journal
for artwork. J,aurelle Shriver for statistical
REFERENCES
1. Brooker .I%, Alderman EL, Harrison DC: Alterations in left ventricular volumes induced by Valsalva maneuver. Br Heart J 36:713. 1974. 2. (irose R. Masking C, Spindola-Franc0 H, Yipintsoi T: Production of left ventricular cavity obliteration in normal man. Circulation 64:448, 1981. 3. Dodge HT, Sandler H, Ballew DW, Lord JD dr: The use 1)1 biplane angiocardiography for measurement of left ventricular volume-in man. AM HEART J 60:762, 1960. 4. Greene DG. Carlisle R. Grant C. Bunneil IL: Estimation of left ventricular volume~by one-plane cineangiography. Circulation 351. 1967. 51 Rackley CE: Quantitative evaluation of left ventricular func tion by radiographic techniques. Circulation 54:862, 1976. 6. Bodenheimer MM, Banka VS, Hermann GA: Reversible asynergy: Histopathologic and electrographic correlations in patients with coronary artery disease. Circulation 53:792, 1976. 7. Ideker RE, Behar VS, Wagner GS. Starr ,JW. Starmer CF. Lee KL. Hackel DB: Evaluation of asvnerev as an indicator of myocardial fibrosis. Circulation 57:715. 1977. 8. Massie BM. Botvinick EH, Brundage BH, Greenberg B, Shames D, Gelberg H: Relationship of regional myocardial perfusion to segmental wall motion. Circulation 58:1154, 1978. 9. Dyke SH, Cohn PF, Gorlin R, Sonnenblick ED: Detection of’ residual myocardial function in coronary artery disease using post-extra systolic potentiation. Circulation 50:694, 1974. LE. Cohn PF. Herman MV. Gorlin R: 10. Horn HR. Teichholz Augmentation of left ventricular contraction pat.tern in corenary artery disease by an inotropic catecholamine. Circulation 49:1063, 1974. 11. Helfant RH. Pine R, Meister SG, Feldman MS, Trout KG. Banka VS: Nitroglycerin to unmask reversible asynergy. Circulation 50:108, 1974. 12. Parisi AF, Harrington JJ, Askenazi J, Pratt RC. McIntyre KM: Echocardiographic evaluation of the Valsalva maneuver in healthy subjects and patients with and without heart failure. Circulation 54:921, 1976. 13. Robertson D, Stevens RM, Friesinger (;c’, Gates JA: The effect of the Valsalva maneuver on echocardiographic dimensions in man. Circulation 55596. 1977. 14. Zema Md. Restivo B. SOS T. Sniderman KW. Kline S: Left ventricular dysfunction-bedside Valsalva manoeuver. Br Heart J 44~560, 1980. 1.5. Leighton RF, Wilt SM, Lewis RP: Detection of hypokinesis by a quantitative analysis of left ventricular cineangiograms. Circulation 50:121, 1974. 16. Gelberg Hd, Brundage BH, Glanz S, Parmley WW: Quantitative left ventricular wall motion analysis: A comparison of area, chord and radial methods. Circulation 59:991. 1979. sinus reflex control of coronary blood flow. 17. Feigh EO: Carotid Circ Res 23:223. 1968. 18. Pepine CJ, Wiener I,: Effects of the Valsalva maneuver on myocardial ischemia in patients with coronary artery disease. Circulation 59:1304, 1979.