- Email: [email protected]
Effect of abruptly increased intrathoracic pressure on coronary blood flow velocity in patients
Effect of abruptly increased intrathoracic pressure on coronary blood flow velocity in patients To assess the effects of abruptly increased intrathoracic pressure on coronary blood flow,
arterial pressure, heart rate, and intracoronary Doppler blood flow velocity were measured continuously during cough(s) and again during the four phases of the Valsalva maneuver in 14 patients. Coughing significantly increased the systolic pressure (137 +- 25 to 176 .+ 30 mm Hg), diastolic pressure (72 _+ 10 to 84 _+ 18 mm Hg), and arterial pulse pressure (65 .+ 27 to 92 .+ 35 mm Hg), with no change in heart rate. The mean coronary flow velocity decreased (17 .+ 10 to 14 .+ 12 c m / s e c , p < 0.03). During the Valsalva maneuver, despite marked reduction in the mean arterial pressure during phase III (96 -+ 12 to 68 .+ 14 mm Hg, p < 0.05), the reduction of coronary blood flow velocity did not achieve statistical significance. These data demonstrate that neither type of abrupt physiologic increase in intrathoracic pressure enhances coronary blood flow. Coughing does not improve coronary perfusion pressures or flow velocity, despite marked increases in arterial diastolic pressure. The Valsalva maneuver, for the most part, does not significantly alter coronary blood flow velocity. (AM HEART J 1990;119:863.)
Morton J. Kern, MD, Chalapathirao Gudipati, MD, Satyam Tatineni, MD, Frank Aguirre, MD, Harvey Serota, MD, and Ubeydullah Deligonul, MD. St. Louis, Mo. Abrupt increase in intrathoracic pressure by coughing has been recommended to maintain blood pressure during severe bradycardia or arterial hypotension after angiography, TM and at onetime was thought to clear contrast from the coronary tree J, 5 Serial coughing can maintain cardiac output during ventricular fibrillation or asystole 2-4 due to increases in intrathoracic pressure forcing blood toward the ascending aorta and brachiocephalic vessels. 6 Increased intrathoracic pressure due to thoracic compression as used in cardiopulmonary resuscitation also produces similar hemodynamic effects in the experimental hypotensive animal. 2, 3 Cough-induced increase in aortic pressure during ventricular fibrillation or asystole were thought to result in an increase in coronary perfusion pressures in some patients. Little et al. 6 measured the hemodynamics during coughing in normal subjects and determined that the cough displaced aortic blood volume peripherally without increasing cardiac output during diastole and preventing venous From the Cardiology Division, St. Louis University Hospital. Received for publication Oct. 2, 1989; accepted Nov. 3, 1989. Reprint requests: Morton J. Kern, MD, Director, Cardiac Catheterization Laboratory, St. Louis University Hospital, Box 15250, 3635 Vista Ave., St. Louis, MO 63110. 4/1/18481
outflow, resulting in a decreased coronary perfusion pressure during and after the cough. Coughing also produced similar decreases in coronary perfusion pressure in patients with moderate hypotension after coronary angiography. It has been postulated that if coughing helps clear coronary arteries of contrast media, it does not do so by increasing coronary flow. However, changes in coronary blood flow d u r i n g coughing in the awake human subject have not been examined. The purpose of this study was to examine the changes in coronary flow velocity during brief, abrupt increases in intrathoracic pressure in awake patients. METHODS Study population. Fourteen patients were studied at the
conclusion of routine diagnostic coronary arteriography and left ventriculography. The protocol for measurement of coronary blood flow velocity in patient s was approved by the Human Subjects Committee of the Institutional Review Board. Written informed consent was obtained in all patients. Patients were studied after an overnight fast and were given no cardioactive medications within 12 to 18 hours prior to study. No sublingual or intracoronary nitroglycerin was used prior to measurements of cough and Valsalva maneuver. Diphenhydramine (25 to 50 mg) and diazepam (5 mg) were given orally as premedication prior 863
864
April 1990 American Heart Journal
Kern et a/.
1 sec
~,
A ECG - -
"-.h,.-"~
__~
200 m m H g l. 100 cm/sec J~
~
~
-
,,
/
Ao-,%.
/,..
____.. "--J
/%
"
~/
/
I~1
sCOwl
\ ',~.J
r% ~,d
~
,-",.
-~
--C
ECG ~
Ao-=,~ k T~ i 1/ [l\k fl~\ A~'/\\ ~\[]J]
I ~ul/'~/"//l/"J L V ~ , ~ J [ / Il l llll -A l l/ lIB/ IIII ~_J._ll I[ I IIr RA
*
,
I
/'-4/I "~
/ ]l l/ fl "
"
,("l~\l\dl /,, I A ~IB'I '~lUfis'~.'A\t ,~\ /"/Ill ~ ~1P'/",r
^ ,,A ~
Fig. 1. A, This panel shows the mean and phasic coronary velocity during a single cough. PV, phasic velocity; MV, mean velocity (centimeters per second). The coronary perfusion pressure (A) is the aortic diastolic - right atrial pressure. The perfusion pressure during cough (B) is the arterial pressure - right atrial pressure. Coronary flow velocity falls in the immediate post-tussive beats. B, This panel shows hemodynamics during cough with the right atrial pressure (RA), left ventricular pressure (L V), aortic pressure (Ao), and electrocardiogram (ECG). The pulse pressure (A) is systolic - diastolic pressure, The perfusion pressure (B) is the aortic diastolic minus left ventricular end-diastolic pressure, or can be measured as (C) the diastolic minus the right atrial mean pressure. During cough, the pulse pressure (A 1) increases, but the perfusion pressures (131) decrease or remain unchanged (C1). In the lower panel, multiple coughs demonstrate the perfusion pressure (B 1) changes. B 1 during any cough is markedly less than during the resting phase.
to diagnostic angiography. All patients received 5000 units of heparin intravenously prior to coronary flow velocity measurements. The study was performed in two parts: hemodynamic and coronary flow velocity measurements were obtained (1) during single and multiple coughs and (in the same patients) (2) during the four phases of the Valsalva maneuver.
Hemodynamic and coronary flow velocity techniques. Phasic aortic pressure was measured through the fluidfilled lumen of an 8F guiding catheter (Shiley Inc., Irvine, Calif.) using a Bentley 900 series transducer (Bentley Laboratories Inc., Irving Calif.). A 2.5F intracoronary Doppler (20 MHz) velocity catheter (Millar Instruments, Houston, Texas) was carefully advanced through the guiding cathe-
Volume
119
Intrathoracic pressure and coronary flow
Number 4
865
1 sec
#,
200 100,
I
I
I
I I I 4 - =_l_.._j..~,L
I
I -, -
I i
I 4_
i
I _&.~ i
i
ic ,4i \
I
\ [ \ ['X ['X ( % I N x4 :x4 ix4 -,J x4
x,~
T
12[
C
~" "
~/I
sec
I
I
I
I
1
I
I
I
I
I
/
ECG200 m m H g ~
loo~n/secA
......
I
+
+
C
/1 ~.-/~--~ /1 a ~1 .~Uk/~ p V ~ . . .
-
I
I
I
I
, r ~
lIT
[ v\
.
.
I
I
I
I
~ .......
I
~ .
.
I
....................
;
I
.
I
.
I
!
.
.
.
.
.
I
.
.
.
I
.
.
.
.
I
:
.
I
~
i
Y ~:~ V~ ~ ~ - 4 ' ~
~i
:
[
I !
! . . . . .
+
7:-T-~7_~:L--_.7:-L:: 5
I
::::-::-
Ao§
Ill
I
..... C ~ .............
:::::::::::::::::::::
~1L l_t
I
/
i:
: ......
---17]
117 i::
-~_~:-:;--~'-~-~.- ~-"i-~--~- ~- :-" ;-- - ,~ , . . . . .
Fig. 2. D a t a records in a p a t i e n t during single (panel A) and multiple (panel B) coughs (C). Duplicate tracings in each panel are obtained from proximal (upper) and more distal (lower) locations within the left coronary artery. In patients in whom forceful coughing is maintained over several seconds, post-tussive hemodynamics demonstrate m a r k e d reduction in coronary flow velocity and possible reversal of coronary flow from the ostial location of velocity measured (top part, lower panel).
ter over a 0.014-inch flexible angioplasty guide wire to the proximal portion of the left anterior descending coronary artery. This technique has been widely u s e d in similar studies. 7,8 Coronary velocity (phasic and mean) signals, electrocardiogram, and arterial pressure were continuously recorded on an optical physiologic recorder (Electronics for Medicine, VR-12, P P G Biomedical Systems Inc., Medical Electronics Division, Pleasantville, N.Y.) at p a p e r speeds of 10 and 100 mm/sec. H e a r t rate was measured from a continuous lead II or V1 electrocardiogram. During cough, hemodynamic d a t a were measured at
peak diastolic right atrial pressure or peak aortic pressure (Fig. 1). The diastolic period of the cough was defined as the time from the dicrotic aortic notch, or the end of E C G T wave until the onset of the QRS complex of the subsequent beat. The systolic pressure was measured at peak aortic pressure during coughing. Coronary velocity was computed at the lowest value for the five consecutive beats after the cough. Representative examples of d a t a records during single and multiple coughs are shown in Fig. 2. The Valsalva maneuver was performed by asking the p a t i e n t to inhale, then strain without exhaling against a
April 1 9 9 0
866
Kern
et al.
American Heart Journal
Fig, 3. D a t a record during the four (I to IV) phases of the Va[salva maneuver. Right atrial pressure (RA) is shown on the top panel (A). Alterations in mean or phasic coronary flow velocity have been observed during the m a r k e d reduction in pressure of Valsalva phase III. When comparing different locations of coronary arterial flow during the strain phases, both the mor e proximal (B, upper tracing) and distal locations (B, lower tracing) show a decrease in coronary flow velocity.
closed glottis. This maneuver was carried out until a fall in arterial pressure was observed. T h e four phases of the Valsalva maneuver were easily recognized (Fig. 3) and d a t a were recorded continously at rest (phase I), during strain (phase II), during pressure decline (phase III), and on release (phase IV). Subjects with abnormalities other than coronary artery disease affecting coronary vasodilatory capacity were excluded from the study. P a t i e n t s excluded were those considered to have recent myocardial infarction by clinical or electrocardiographic criteria, and marked left ventricular hypertrophy. One patient (No. 7) with severe left ventricular dysfunction due to nonischemic cardiomyopathy was included. P a t i e n t s with autonomic dysfunction, such as diabetes mellitus, orthostatic hypertension, Parkinson's disease, or renal insufficiency were also not enrolled in this
protocol. Measurement of coronary vasodilatory reserve. Coronary vasodilatory reserve was assessed by the intracoronary blood flow velocity responses to the administration of papaverine. 7 Measurement of resting coronary blood flow velocity was o b t a i n e d and 10 mg of papaverine hydrochloride in 10 ml 3 normal saline was injected through the guiding catheter into the coronary artery and maximal coronary blood flow velocity was recorded. Maximal coronary hyperemia was confirmed by a second 2 mg higher dose o f p a paverine in four patients. CorOnary flow velocity was
allowed to return to baseline between doses of papaverine. This technique has been previously demonstrated to produce maximal coronary hyperemia and has been used in the assessment of flow reserve for patients with significant coronary stenosis both before and after angioplasty, s Data analysis. Coronary vasodilatory reserve was calculated as the quotient of peak mean flow velocity to resting blood flow velocity. Phasic and mean velocity values were obtained immediately before coughing and the maximal change was obtained immediately after coughing and during maximal decrease in aortic pressure during phase III of the Valsalva maneuver. Coronary resistance was calculated as mean arterial pressure/mean coronary velocity. Peak phasic and mean coronary flow velocities (centimeters per second) were calibrated from an arbitrarily set internal 0 to 100 cm/sec full scale. T h e reproducibility of these measures is < _+ 10%. Mean arterial pressure was calculated from the phasic arterial wave form. Pulse pressure was c o m p u t e d as aortic systolic minus aortic diastolic pressure. Statistical analysis. For the Valsalva data, values were analyzed by analysis of variance for multiple comparisons. W h e n a significant difference indicated contribution to sample variability, Scheff~'s test was performed. For the cough d a t a with paired values , S t u d e n t ' s t test was used. A probability (p) value of <0.05 was considered statistically significant. Results are expressed as mean _+ 1 s t a n d a r d deviation.
V o l u m e 119
Intrathoracic p r e s s u r e a n d coronary flow
Number 4
867
Table I. H e m o d y n a m i c characteristics Patient No.
Age
Sex
L V E F (%)
L V E D P (mm Hg)
CAD (%)
CVR (Units)
1 2 3 4 5 6 7 8 9 10 11 12 13 14
72 51 60 77 36 65 71 46 56 59 59 60 54 69 60 • 11
M F M F M F M M M M M M M M
66 50 73 84 64 50 30 50 69 71 70 72 65 82 64 • 14
5 5 6 0 18 12 15 5 20 14 8 10 16 30 12 • 8
L A D 80, R C A 50 -R C A 90 -----C F X 8 0 (distal) --C F X 8 5 (distal) ---
2.02 1.86 2.15 1.80 2.10 1.77 1.54 2.15 2.23 3.66 4.64 2.78 2.92 1.48 2.48 • 0.95
LVEF, Left ventricular ejection fraction; LVEDP, left ventricular end-diastolic pressure; CAD, coronary artery disease, percent diameter narrowing; LAD, left anterior descending artery; RCA, right coronary artery; CFX, circumflex coronary artery; CVR, coronary vasodilatory reserve.
Table IIA. V a l s a l v a and cough d a t a Valsalva phase I Systo lic ( m m Hg) D i a s t o l i c ( m m Hg) M e a n a r t e r i a l ( m m Hg) P u l s e p r e s s u r e (ram Hg) Heart rate (beats/min) M e a n v e l o c i t y (cm/sec) Peak phasic velocity (cm/sec) CR (units)
133 78 96 55 75 15 25
• • + + + + +
H 22 15 14 9 14 10 14
9 • 6
150 96 114 53 69 14 23
+ 28 • 17 _+ 16 • 28 • 16 _+ 8 + 14
11 _+ 8
Cough
III 83 61 68 22 80 11 19
IV
• 13" _+ 12" • 11" _+ 9:~ _+ 18 _+ 9 + 15
13 + 15
138 79 99 59 74 13 21
• 19 _ 12 -+ 13 • 13 • 20 • 8 _+ 12
10 _+ 7
Control 133 74 94 59 76 17 30
Cough
+ 22 + 9 -+ 9 _+ 25 • 8 • 10 • 14
173 87 116 86 77 14 26
7 _+ 4
• + -+ + • + •
% Aa
29]; 17~/ 155 335 13 12t 17t
31 19 24 71 2 -21 -15
19 _+ 275
• 20 _+ 24 • 13 • 110 • 7 • 31 • 30
150 • 350
%Aa , Change during each phase of Valsalva maneuver or during cough (see Table IIB for relative changes during Valsalva maneuvers); CR, coronary resistance (mean arterial pressure/mean velocity). *p < 0.05, tP < 0.03, :~p < 0.01 versus control or Valsalva phase II.
RESULTS
Clinical characteristic. There were 11 men and 3 women, with a mean age of 60 _+ 11 years (Table I).
Left ventricular function was normal in all but one patient (No. 7) with a compensated idiopathic cardiomyopathy. Limited coronary disease was present in four patients with mid left anterior descending (No. 1), right coronary artery (No. 3), and distal circumflex disease (Nos. 9 and 12), respectively. Patient No. 4 had prior left anterior descending percutaneous transluminal coronary angioplasty (PTCA) (>1 year) with no residual narrowing. Cough (Table IIA). During cough, systolic, diastolic, and pulse pressure were all increased significantly (Fig. 4, A), from 133 + 22 to 173 _+ 29 mm Hg (p < 0.01), 74 + 9 to 87 _+ 17 mm Hg (p < 0.01), and 59 _+ 25 to 86 +_ 33 mm Hg (p < 0.01), respectively, with no changes in heart rate. Both the mean velocity (from 17 +_ 10 to 14 _+ 12, p < 0.03) and peak
Table liB. Percent change during Valsalva maneuver Phase
S ys t ol i c (ram Hg) Diastolic (mm Hg) Mean arterial pressure ( m m Hg) P u l s e pre s s ure ( m m Hg) Heart rate (beats/rain) M e a n velocity (cm/sec) P e a k p h a s i c velocity (cm/sec)
I vs H
H vs I I I
I I I vs I V
13 +_ 16 27 • 28 20 • 20
- 4 4 • 10 - 3 4 • 20 - 3 9 • 13
69 • 24 32 + 16 47 _+ 18
-5 -7 -10 -15
• 37 • 10 _+ 36 +_ 35
-55 17 -24 -21
+ 12 • 12 _+ 29 + 24
202 -8 66 46
_+ 116 _+ 10 _ 111 • 72
phasic coronary flow velocity (from 30 + 14 to 26 _+ 17 cm/sec, p < 0.03) decreased during cough. Coronary resistance increased (7 _+ 4 to 19 +_ 27 units, p < 0.01) during cough.
April
868
A
Kern et al.
83 + 13 m m Hg (p < 0.05) and 78 _+ 15 to 61 + 12 mm H g ( p < 0.05), respectively, returning to control values during phase IV (Fig. 4, B). Heart rate tended to increase during the end of the phase III period. The mean coronary flow velocity remained unchanged at 15 _+ 10 cm/sec, 14 + 8 cm/sec, 11 + 9 cm/sec, and 13 _+ 8 cm/sec for the four phases of the Valsalva maneuver, respectively.
Cough and Coronary Velocity
200
T
150
1990
American Heart Journal
i "~
I
DISCUSSION
100
50
i, v [ - - -
o
1
1
T
~--
!
9--
ll~
9 DIA
I
i
,svs
I
I
Control
B
opv #
oHR
Cough
Valsalva and Coronary Velocity
200,
II
I
l
IV
150
"l 1 5 ! ! ~
ii, X I
100
;-----f "
opv gMV oHR IDIA
50
-SYS i
Phase 1
I
Phase 2
I
Phase 3
I
Phase4
Fig. 4. Hemodynamic responses during cough (A). Peak phasic velocity (PV, centimeters per second), mean velocity (MV, centimeters per second), heart rate (HR, beats per minute), diastolic pressure (DIA, millimeters of mercury), and systolic pressure (SYS, millimeters of mercury) are shown. *p < 0.03. Scale is in 200 mm Hg, beats per minute, and centimeters per second for pressure, heart rate, and velocity, respectively. Valsalva.maneuver (Tables IIA and HB). For the four phases of the Valsalva maneuver, the significant differences occurred during phase III with a decline in systolic and diastolic pressure from 133 +_ 22 to
The most abrupt increases in intrathoracic pressure decreased coronary blood flow velocity. These findings support a previous hypothesis 6,12 that the decreased perfusion pressure during coughing decreases coronary flow. During cough, the large arteries transmit increased intrathoracic pressure to the periphery, maintaining systemic pressure) -4,9,1~ However, although the gradient for peripheral blood flow during cough is increased, the gradient for coronary blood flow is decreased. 6,10, 12 Aortic pressure increases less than intrathoracic pressure 3, 10,11 or right atrial pressure, and thus coronary perfusion pressure (aortic minus right atrial pressure) decreases. Sustained increases in thoracic pressure (e.g., multiple coughs) lower aortic pressure and coronary blood flow velocity immediately and more so than single brief coughs in these normotensive subjects. The mechanisms of reduced coronary flow velocity reflected both the reduced aortic volume (and pressure) after cough and a possible reversal of left coronary flow due to mechanical compression of the coronary arteries 13 during cough. Transient reversal of coronary blood flow during ventricular systolic compression of the intramyocardial vessel has been demonstrated in patients with aortic stenosis 13 and in experimental animals with acute aortic insufficiency)4 In hypotensive patients, coughing may increase aortic flow.6 The sudden increase in intrathoracic pressure or external thoracic compression displaces peripheral blood from the central aorta and, in contrast to normotensive individuals, produces a gradient for opening of the aortic valve 6 but without a favorable left ventricular transmural gradient. The echocardiographic and hemodynamic demonstration of reduced coronary perfusion pressure by Little et al. 6 did not prove that coughing altered flow in coronary arteries. Reflex mechanisms during cough produce coronary vasodilatation in dogs) 5 However, the pulmonary inflation reflex in man is somewhat different from that in dogs) 6 In man, coronary enlargement during respiratory maneuvers has not been observed. 15 Effect of Valsalva maneuver ity.
and coronary
flow veloc-
The hemodynamics of the Valsalva maneuver in
Volume 119 Number 4
our patients are identical to those classical descriptions in which straining increases intrathoracic pressure abruptly, 17-19 The increased pressure is transmitted to the thoracic aorta with an increase in arterial blood pressure. Increased intrathoracic pressure reduces venous return, producing a decline in right and then in left ventricular end-diastolic volume, resulting in a fall of left ventricular stroke volume and a decline in mean arterial pressure and pulse pressure in phase III of the Valsalva maneuver. 19-21The fall in arterial pressure is accompanied by a reflex increase in heart rate and peripheral vasoconstriction. If autoregulation remains intact, coronary blood flow should decrease during the reduction of determinants of myocardial oxygen demand of the Valsalva maneuver. 23 In the current study, coronary flow velocity tended to decrease, but the decrease was not of the magnitude previously demonstrated. 15, 22 Coronary flow velocity was most significantly affected during phase III and remained relatively constant, adjusting to rapid changes in systemic pressure and heart rate. Benchimol et al., 22 measuring coronary ostial blood flow velocity, found similar changes of greater magnitude during the Valsalva maneuver. There was no difference in this response in patients with and without coronary artery disease. Wilson et a l ) 5 also found decreases in subselective coronary blood flow velocity in normal patients during the Valsalva maneuver. Changes in coronary flow velocity in 12 patients with normal coronary arteries and normal coronary flow reserve (4.8 _+ 0.20 units) during the Valsalva maneuver were different from the responses observed in dogs. Coronary blood flow velocity was reduced 0.67 _+ 0.9 times the resting value with increased coronary resistance (1.37 _+ 0.20 units, p = 0.05 versus control) during the Valsalva maneuver. No significant alterations in coronary blood flow velocity have been reported during deep breathing, the Muller maneuver, yawn, handgrip, and deep breathing during handgrip. 15 In patients, autonomic regulation of the coronary circulation does not appear to play a major role during the pulmonary inflation via reflex activation) 5 The patients in the current study differed from those of Wilson et al. 15 in that impaired coronary reserve was more prevalent and may play more of a role in attenuating the coronary flow response to the Valsalva maneuver. The current findings support previous results in which coronary flow decreases with increases in thoracic pressure. Whether reduced coronary reserve and some degree of coronary atherosclerosis alter coronary flow velocity responses remains under study. Limitations. Abnormalities that might affect the vasodilatory capacity of the arterial vasculature such as myocardial infarction, left ventricular hypertro-
Intrathoracic pressure and coronary flow
869
phy, left ventricular dysfunction, autonomic dysfunction, diabetes mellitus, orthostatic hypertension, Parkinson's disease, and renal dysfunction might produce different findings relative to coronary flow velocity changes in patients with normal coronary vasodilatory reserve. For this study, only one patient with left ventricular dysfunction was examined. Results are unchanged when this patient's responses are excluded. Doppler measurement of coronary blood flow velocity provides relative rather than absolute blood flow. Alterations in cross-sectional area of the artery during respiratory maneuvers or coughing might affect the relationship between flow and flow velocity. Utilizing quantitative angiography, Wilson et al. 15 demonstrated that coronary luminal area during the respiratory maneuvers described here in unchanged. Pharmacologic therapy may also influence the reflex responses. However, in similar patients with and without premedication, no differences were observed. 15Wilson et al) 5 noted a marked increase in coronary resistance during peak Valsalva maneuver, presumably due to sympathetic discharge and fall in coronary resistance during intracoronary papaverine, indicating that the coronary vasculature appeared to be capable of normal vasoconstriction and vasodilatation. Our study findings differ in that a majority of patients had a subnormal (<3.5 units) vasodilatory response. However, similar responses in peak and mean coronary flow velocity occurred for patients with and without coronary vasodilatory reserve >3.0 units. Also, similar coronary flow velocity responses were demonstrated for patients with and without coronary artery disease, 22 indicating little contribution of coronary disease to coronary flow responses to abrupt increases in intrathoracic pressure. Clinical significance. Abrupt increases in intrathoracic pressure (e.g., coughing) do not increase coronary blood flow velocity. Maintenance of arterial pressure for cerebral perfusion appears to be the major benefit during hypotensive conditions. The Valsalva maneuver tends to decrease coronary blood flow and is not significantly influenced by the decreased coronary perfusion pressure in patients with normal or impaired coronary reserve. The authors wish to thank the Mudd Cardiac Catheterization Team, and also Donna Sander for manuscript preparation. REFERENCES
1. Gensini GG. Coronary arteriography. In: Braunwald E, ed. Heart disease: A textbook of cardiovascular medicine. Philadelphia: WB Saunders Co, 1980:309. 2. Rudikoff MT, Maughan WL, Effron M, Freund P, Weisfeldt ML. Mechanisms of blood flow during cardiopulmonary resuscitation., Circulation 1980;61:345. 3. Rosborough JP, Hausknecht M, Criley JM, Garner D, Nie-
870
4. 5. 6.
7.
8.
9. 10. 11. 12.
Kern et at. mann JT. Cardiac output during ventricular fibrillation in the coughing dog [Abstract]. Am J Cardiol 1980;45:445. Sharpey-Schafer EP. Effects of coughing on intrathoracic pressure, arterial pressure and peripheral blood flow. J Physiol 1953;122:351. Pepine CJ. Coronary circulatory effects of increased intrathoracic pressure in intact dogs. Chest 1977;1:72. Little WC, Reeves RC, Coughlan C, Rogers EW. Effect of cough on coronary perfusion pressure: does coughing help clear the coronary arteries of angiographic contrast medium? Circulation 1982;65:604. Wilson RF, Laughlin DE, Ackell PH, Chilian WM, Holida MD, Hartley CJ, Armstrong ML, Marcus ML, White CW. Transluminal, subselective measurement of coronary artery blood flow velocity and vasodilator reserve in man. Circulation 1985;72:82. Kern MJ, Deligonul U, Vandormael M, Labovitz A, Gudipati CV, Gabliani G, Bodet J, Shah Y, Kennedy HL. Impaired coronary vasodilator reserve in the immediate postcoronary angioplasty period: analysis of coronary artery flow velocity indexes and regional cardial venous efflux. J Am Coll Cardiol 1989;13:860. Niemann JT, Garner D, Rosborougb J, Criley JM. The mechanism of blood flow in closed chest cardiopulmonary resuscitation [Abstract]. Circulation 1979;60(supp1 II):II-74. Hamilton WF, Woodbury RA, Harper HT. Arterial, cerebrospinal and venous pressures in man during cough and strain. Am J Physiol 1944;141:42. McIntosh HD, Estes EH, Warren JV. The mechanism of cough syncope. AM HEARTJ 1956;52:70. Cohen A, Gottdiener J, Wish M, Fletcher R. Limitations of cough in maintaining blood flow during asystole: assessment by two-dimensional and Doppler echocardiography. AMHEART J 1989;118:474.
April 1990 American Heart Journal
13. Carroll RJ, Falsetti HL. Retrograde coronary artery flow in aortic valve disease. Circulation 1976;54:494. 14. Folts JD, Rowe GG. Coronary and hemodynamic effects of temporary acute aortic insufficiency in intact anesthetized dogs. Circ Res 1974;35:238. 15. Wilson RF, Marcus ML, White CW. Pulmonary inflation reflex: its lack of physiological significance in coronary circulation of humans. Am J Physiol 1988;255(Heart Circ Physiol 24):H866. 16. Vatner SF, McRitchie RJ. Interaction of the chemoreflex and the pulmonary inflation reflex in the regulation of coronary circulation in conscious dogs. Circ Res 1975;37:664. 17. Hamilton WF, Woodbury RA, Harper HT Jr. Arterial cerebrospinal and venous pressures in man during cough and strain. Am J Physiol 1944;141:42. 18. Sharpey-Schafer EP. Effects of Valsalva's manoeuvre on the normal and failing circulation. Br Med J 1955;2:693. 19. Buda AJ, Pinsky MR, Ingels NB, Doughters GT, Stinson EB, Alderman EL. Effect of intrathoracic pressure on left ventricular performance. N Engl J Med 1979;301:453. 20. Brooker JZ, Alderman EL, Harrison DC. Alterations in left ventricular volumes induced by Valsalva manoeuvre. Br Heart J 1974;36:713. 21. Greenfield JC, Cox RL, Hernandex RR, Thomas C, Schoonmaker FW: Pressure-flow studies in man during the Valsalva maneuver with observations on the mechanical properties of the ascending aorta. Circulation 1967;35:653. 22. Benchimol A, Wang TF, Desser KB, Gartlan JL Jr. The Valsalva maneuver and coronary arterial blood flow velocity. Ann Intern Med 1972;77:357. 23. Pepine CJ, Wiener L. Effects of the Valsalva maneuver on myocardial ischemia in patients with coronary artery disease. Circulation 1979;59:1304.
arterial pressure, heart rate, and intracoronary Doppler blood flow velocity were measured continuously during cough(s) and again during the four phases of the Valsalva maneuver in 14 patients. Coughing significantly increased the systolic pressure (137 +- 25 to 176 .+ 30 mm Hg), diastolic pressure (72 _+ 10 to 84 _+ 18 mm Hg), and arterial pulse pressure (65 .+ 27 to 92 .+ 35 mm Hg), with no change in heart rate. The mean coronary flow velocity decreased (17 .+ 10 to 14 .+ 12 c m / s e c , p < 0.03). During the Valsalva maneuver, despite marked reduction in the mean arterial pressure during phase III (96 -+ 12 to 68 .+ 14 mm Hg, p < 0.05), the reduction of coronary blood flow velocity did not achieve statistical significance. These data demonstrate that neither type of abrupt physiologic increase in intrathoracic pressure enhances coronary blood flow. Coughing does not improve coronary perfusion pressures or flow velocity, despite marked increases in arterial diastolic pressure. The Valsalva maneuver, for the most part, does not significantly alter coronary blood flow velocity. (AM HEART J 1990;119:863.)
Morton J. Kern, MD, Chalapathirao Gudipati, MD, Satyam Tatineni, MD, Frank Aguirre, MD, Harvey Serota, MD, and Ubeydullah Deligonul, MD. St. Louis, Mo. Abrupt increase in intrathoracic pressure by coughing has been recommended to maintain blood pressure during severe bradycardia or arterial hypotension after angiography, TM and at onetime was thought to clear contrast from the coronary tree J, 5 Serial coughing can maintain cardiac output during ventricular fibrillation or asystole 2-4 due to increases in intrathoracic pressure forcing blood toward the ascending aorta and brachiocephalic vessels. 6 Increased intrathoracic pressure due to thoracic compression as used in cardiopulmonary resuscitation also produces similar hemodynamic effects in the experimental hypotensive animal. 2, 3 Cough-induced increase in aortic pressure during ventricular fibrillation or asystole were thought to result in an increase in coronary perfusion pressures in some patients. Little et al. 6 measured the hemodynamics during coughing in normal subjects and determined that the cough displaced aortic blood volume peripherally without increasing cardiac output during diastole and preventing venous From the Cardiology Division, St. Louis University Hospital. Received for publication Oct. 2, 1989; accepted Nov. 3, 1989. Reprint requests: Morton J. Kern, MD, Director, Cardiac Catheterization Laboratory, St. Louis University Hospital, Box 15250, 3635 Vista Ave., St. Louis, MO 63110. 4/1/18481
outflow, resulting in a decreased coronary perfusion pressure during and after the cough. Coughing also produced similar decreases in coronary perfusion pressure in patients with moderate hypotension after coronary angiography. It has been postulated that if coughing helps clear coronary arteries of contrast media, it does not do so by increasing coronary flow. However, changes in coronary blood flow d u r i n g coughing in the awake human subject have not been examined. The purpose of this study was to examine the changes in coronary flow velocity during brief, abrupt increases in intrathoracic pressure in awake patients. METHODS Study population. Fourteen patients were studied at the
conclusion of routine diagnostic coronary arteriography and left ventriculography. The protocol for measurement of coronary blood flow velocity in patient s was approved by the Human Subjects Committee of the Institutional Review Board. Written informed consent was obtained in all patients. Patients were studied after an overnight fast and were given no cardioactive medications within 12 to 18 hours prior to study. No sublingual or intracoronary nitroglycerin was used prior to measurements of cough and Valsalva maneuver. Diphenhydramine (25 to 50 mg) and diazepam (5 mg) were given orally as premedication prior 863
864
April 1990 American Heart Journal
Kern et a/.
1 sec
~,
A ECG - -
"-.h,.-"~
__~
200 m m H g l. 100 cm/sec J~
~
~
-
,,
/
Ao-,%.
/,..
____.. "--J
/%
"
~/
/
I~1
sCOwl
\ ',~.J
r% ~,d
~
,-",.
-~
--C
ECG ~
Ao-=,~ k T~ i 1/ [l\k fl~\ A~'/\\ ~\[]J]
I ~ul/'~/"//l/"J L V ~ , ~ J [ / Il l llll -A l l/ lIB/ IIII ~_J._ll I[ I IIr RA
*
,
I
/'-4/I "~
/ ]l l/ fl "
"
,("l~\l\dl /,, I A ~IB'I '~lUfis'~.'A\t ,~\ /"/Ill ~ ~1P'/",r
^ ,,A ~
Fig. 1. A, This panel shows the mean and phasic coronary velocity during a single cough. PV, phasic velocity; MV, mean velocity (centimeters per second). The coronary perfusion pressure (A) is the aortic diastolic - right atrial pressure. The perfusion pressure during cough (B) is the arterial pressure - right atrial pressure. Coronary flow velocity falls in the immediate post-tussive beats. B, This panel shows hemodynamics during cough with the right atrial pressure (RA), left ventricular pressure (L V), aortic pressure (Ao), and electrocardiogram (ECG). The pulse pressure (A) is systolic - diastolic pressure, The perfusion pressure (B) is the aortic diastolic minus left ventricular end-diastolic pressure, or can be measured as (C) the diastolic minus the right atrial mean pressure. During cough, the pulse pressure (A 1) increases, but the perfusion pressures (131) decrease or remain unchanged (C1). In the lower panel, multiple coughs demonstrate the perfusion pressure (B 1) changes. B 1 during any cough is markedly less than during the resting phase.
to diagnostic angiography. All patients received 5000 units of heparin intravenously prior to coronary flow velocity measurements. The study was performed in two parts: hemodynamic and coronary flow velocity measurements were obtained (1) during single and multiple coughs and (in the same patients) (2) during the four phases of the Valsalva maneuver.
Hemodynamic and coronary flow velocity techniques. Phasic aortic pressure was measured through the fluidfilled lumen of an 8F guiding catheter (Shiley Inc., Irvine, Calif.) using a Bentley 900 series transducer (Bentley Laboratories Inc., Irving Calif.). A 2.5F intracoronary Doppler (20 MHz) velocity catheter (Millar Instruments, Houston, Texas) was carefully advanced through the guiding cathe-
Volume
119
Intrathoracic pressure and coronary flow
Number 4
865
1 sec
#,
200 100,
I
I
I
I I I 4 - =_l_.._j..~,L
I
I -, -
I i
I 4_
i
I _&.~ i
i
ic ,4i \
I
\ [ \ ['X ['X ( % I N x4 :x4 ix4 -,J x4
x,~
T
12[
C
~" "
~/I
sec
I
I
I
I
1
I
I
I
I
I
/
ECG200 m m H g ~
loo~n/secA
......
I
+
+
C
/1 ~.-/~--~ /1 a ~1 .~Uk/~ p V ~ . . .
-
I
I
I
I
, r ~
lIT
[ v\
.
.
I
I
I
I
~ .......
I
~ .
.
I
....................
;
I
.
I
.
I
!
.
.
.
.
.
I
.
.
.
I
.
.
.
.
I
:
.
I
~
i
Y ~:~ V~ ~ ~ - 4 ' ~
~i
:
[
I !
! . . . . .
+
7:-T-~7_~:L--_.7:-L:: 5
I
::::-::-
Ao§
Ill
I
..... C ~ .............
:::::::::::::::::::::
~1L l_t
I
/
i:
: ......
---17]
117 i::
-~_~:-:;--~'-~-~.- ~-"i-~--~- ~- :-" ;-- - ,~ , . . . . .
Fig. 2. D a t a records in a p a t i e n t during single (panel A) and multiple (panel B) coughs (C). Duplicate tracings in each panel are obtained from proximal (upper) and more distal (lower) locations within the left coronary artery. In patients in whom forceful coughing is maintained over several seconds, post-tussive hemodynamics demonstrate m a r k e d reduction in coronary flow velocity and possible reversal of coronary flow from the ostial location of velocity measured (top part, lower panel).
ter over a 0.014-inch flexible angioplasty guide wire to the proximal portion of the left anterior descending coronary artery. This technique has been widely u s e d in similar studies. 7,8 Coronary velocity (phasic and mean) signals, electrocardiogram, and arterial pressure were continuously recorded on an optical physiologic recorder (Electronics for Medicine, VR-12, P P G Biomedical Systems Inc., Medical Electronics Division, Pleasantville, N.Y.) at p a p e r speeds of 10 and 100 mm/sec. H e a r t rate was measured from a continuous lead II or V1 electrocardiogram. During cough, hemodynamic d a t a were measured at
peak diastolic right atrial pressure or peak aortic pressure (Fig. 1). The diastolic period of the cough was defined as the time from the dicrotic aortic notch, or the end of E C G T wave until the onset of the QRS complex of the subsequent beat. The systolic pressure was measured at peak aortic pressure during coughing. Coronary velocity was computed at the lowest value for the five consecutive beats after the cough. Representative examples of d a t a records during single and multiple coughs are shown in Fig. 2. The Valsalva maneuver was performed by asking the p a t i e n t to inhale, then strain without exhaling against a
April 1 9 9 0
866
Kern
et al.
American Heart Journal
Fig, 3. D a t a record during the four (I to IV) phases of the Va[salva maneuver. Right atrial pressure (RA) is shown on the top panel (A). Alterations in mean or phasic coronary flow velocity have been observed during the m a r k e d reduction in pressure of Valsalva phase III. When comparing different locations of coronary arterial flow during the strain phases, both the mor e proximal (B, upper tracing) and distal locations (B, lower tracing) show a decrease in coronary flow velocity.
closed glottis. This maneuver was carried out until a fall in arterial pressure was observed. T h e four phases of the Valsalva maneuver were easily recognized (Fig. 3) and d a t a were recorded continously at rest (phase I), during strain (phase II), during pressure decline (phase III), and on release (phase IV). Subjects with abnormalities other than coronary artery disease affecting coronary vasodilatory capacity were excluded from the study. P a t i e n t s excluded were those considered to have recent myocardial infarction by clinical or electrocardiographic criteria, and marked left ventricular hypertrophy. One patient (No. 7) with severe left ventricular dysfunction due to nonischemic cardiomyopathy was included. P a t i e n t s with autonomic dysfunction, such as diabetes mellitus, orthostatic hypertension, Parkinson's disease, or renal insufficiency were also not enrolled in this
protocol. Measurement of coronary vasodilatory reserve. Coronary vasodilatory reserve was assessed by the intracoronary blood flow velocity responses to the administration of papaverine. 7 Measurement of resting coronary blood flow velocity was o b t a i n e d and 10 mg of papaverine hydrochloride in 10 ml 3 normal saline was injected through the guiding catheter into the coronary artery and maximal coronary blood flow velocity was recorded. Maximal coronary hyperemia was confirmed by a second 2 mg higher dose o f p a paverine in four patients. CorOnary flow velocity was
allowed to return to baseline between doses of papaverine. This technique has been previously demonstrated to produce maximal coronary hyperemia and has been used in the assessment of flow reserve for patients with significant coronary stenosis both before and after angioplasty, s Data analysis. Coronary vasodilatory reserve was calculated as the quotient of peak mean flow velocity to resting blood flow velocity. Phasic and mean velocity values were obtained immediately before coughing and the maximal change was obtained immediately after coughing and during maximal decrease in aortic pressure during phase III of the Valsalva maneuver. Coronary resistance was calculated as mean arterial pressure/mean coronary velocity. Peak phasic and mean coronary flow velocities (centimeters per second) were calibrated from an arbitrarily set internal 0 to 100 cm/sec full scale. T h e reproducibility of these measures is < _+ 10%. Mean arterial pressure was calculated from the phasic arterial wave form. Pulse pressure was c o m p u t e d as aortic systolic minus aortic diastolic pressure. Statistical analysis. For the Valsalva data, values were analyzed by analysis of variance for multiple comparisons. W h e n a significant difference indicated contribution to sample variability, Scheff~'s test was performed. For the cough d a t a with paired values , S t u d e n t ' s t test was used. A probability (p) value of <0.05 was considered statistically significant. Results are expressed as mean _+ 1 s t a n d a r d deviation.
V o l u m e 119
Intrathoracic p r e s s u r e a n d coronary flow
Number 4
867
Table I. H e m o d y n a m i c characteristics Patient No.
Age
Sex
L V E F (%)
L V E D P (mm Hg)
CAD (%)
CVR (Units)
1 2 3 4 5 6 7 8 9 10 11 12 13 14
72 51 60 77 36 65 71 46 56 59 59 60 54 69 60 • 11
M F M F M F M M M M M M M M
66 50 73 84 64 50 30 50 69 71 70 72 65 82 64 • 14
5 5 6 0 18 12 15 5 20 14 8 10 16 30 12 • 8
L A D 80, R C A 50 -R C A 90 -----C F X 8 0 (distal) --C F X 8 5 (distal) ---
2.02 1.86 2.15 1.80 2.10 1.77 1.54 2.15 2.23 3.66 4.64 2.78 2.92 1.48 2.48 • 0.95
LVEF, Left ventricular ejection fraction; LVEDP, left ventricular end-diastolic pressure; CAD, coronary artery disease, percent diameter narrowing; LAD, left anterior descending artery; RCA, right coronary artery; CFX, circumflex coronary artery; CVR, coronary vasodilatory reserve.
Table IIA. V a l s a l v a and cough d a t a Valsalva phase I Systo lic ( m m Hg) D i a s t o l i c ( m m Hg) M e a n a r t e r i a l ( m m Hg) P u l s e p r e s s u r e (ram Hg) Heart rate (beats/min) M e a n v e l o c i t y (cm/sec) Peak phasic velocity (cm/sec) CR (units)
133 78 96 55 75 15 25
• • + + + + +
H 22 15 14 9 14 10 14
9 • 6
150 96 114 53 69 14 23
+ 28 • 17 _+ 16 • 28 • 16 _+ 8 + 14
11 _+ 8
Cough
III 83 61 68 22 80 11 19
IV
• 13" _+ 12" • 11" _+ 9:~ _+ 18 _+ 9 + 15
13 + 15
138 79 99 59 74 13 21
• 19 _ 12 -+ 13 • 13 • 20 • 8 _+ 12
10 _+ 7
Control 133 74 94 59 76 17 30
Cough
+ 22 + 9 -+ 9 _+ 25 • 8 • 10 • 14
173 87 116 86 77 14 26
7 _+ 4
• + -+ + • + •
% Aa
29]; 17~/ 155 335 13 12t 17t
31 19 24 71 2 -21 -15
19 _+ 275
• 20 _+ 24 • 13 • 110 • 7 • 31 • 30
150 • 350
%Aa , Change during each phase of Valsalva maneuver or during cough (see Table IIB for relative changes during Valsalva maneuvers); CR, coronary resistance (mean arterial pressure/mean velocity). *p < 0.05, tP < 0.03, :~p < 0.01 versus control or Valsalva phase II.
RESULTS
Clinical characteristic. There were 11 men and 3 women, with a mean age of 60 _+ 11 years (Table I).
Left ventricular function was normal in all but one patient (No. 7) with a compensated idiopathic cardiomyopathy. Limited coronary disease was present in four patients with mid left anterior descending (No. 1), right coronary artery (No. 3), and distal circumflex disease (Nos. 9 and 12), respectively. Patient No. 4 had prior left anterior descending percutaneous transluminal coronary angioplasty (PTCA) (>1 year) with no residual narrowing. Cough (Table IIA). During cough, systolic, diastolic, and pulse pressure were all increased significantly (Fig. 4, A), from 133 + 22 to 173 _+ 29 mm Hg (p < 0.01), 74 + 9 to 87 _+ 17 mm Hg (p < 0.01), and 59 _+ 25 to 86 +_ 33 mm Hg (p < 0.01), respectively, with no changes in heart rate. Both the mean velocity (from 17 +_ 10 to 14 _+ 12, p < 0.03) and peak
Table liB. Percent change during Valsalva maneuver Phase
S ys t ol i c (ram Hg) Diastolic (mm Hg) Mean arterial pressure ( m m Hg) P u l s e pre s s ure ( m m Hg) Heart rate (beats/rain) M e a n velocity (cm/sec) P e a k p h a s i c velocity (cm/sec)
I vs H
H vs I I I
I I I vs I V
13 +_ 16 27 • 28 20 • 20
- 4 4 • 10 - 3 4 • 20 - 3 9 • 13
69 • 24 32 + 16 47 _+ 18
-5 -7 -10 -15
• 37 • 10 _+ 36 +_ 35
-55 17 -24 -21
+ 12 • 12 _+ 29 + 24
202 -8 66 46
_+ 116 _+ 10 _ 111 • 72
phasic coronary flow velocity (from 30 + 14 to 26 _+ 17 cm/sec, p < 0.03) decreased during cough. Coronary resistance increased (7 _+ 4 to 19 +_ 27 units, p < 0.01) during cough.
April
868
A
Kern et al.
83 + 13 m m Hg (p < 0.05) and 78 _+ 15 to 61 + 12 mm H g ( p < 0.05), respectively, returning to control values during phase IV (Fig. 4, B). Heart rate tended to increase during the end of the phase III period. The mean coronary flow velocity remained unchanged at 15 _+ 10 cm/sec, 14 + 8 cm/sec, 11 + 9 cm/sec, and 13 _+ 8 cm/sec for the four phases of the Valsalva maneuver, respectively.
Cough and Coronary Velocity
200
T
150
1990
American Heart Journal
i "~
I
DISCUSSION
100
50
i, v [ - - -
o
1
1
T
~--
!
9--
ll~
9 DIA
I
i
,svs
I
I
Control
B
opv #
oHR
Cough
Valsalva and Coronary Velocity
200,
II
I
l
IV
150
"l 1 5 ! ! ~
ii, X I
100
;-----f "
opv gMV oHR IDIA
50
-SYS i
Phase 1
I
Phase 2
I
Phase 3
I
Phase4
Fig. 4. Hemodynamic responses during cough (A). Peak phasic velocity (PV, centimeters per second), mean velocity (MV, centimeters per second), heart rate (HR, beats per minute), diastolic pressure (DIA, millimeters of mercury), and systolic pressure (SYS, millimeters of mercury) are shown. *p < 0.03. Scale is in 200 mm Hg, beats per minute, and centimeters per second for pressure, heart rate, and velocity, respectively. Valsalva.maneuver (Tables IIA and HB). For the four phases of the Valsalva maneuver, the significant differences occurred during phase III with a decline in systolic and diastolic pressure from 133 +_ 22 to
The most abrupt increases in intrathoracic pressure decreased coronary blood flow velocity. These findings support a previous hypothesis 6,12 that the decreased perfusion pressure during coughing decreases coronary flow. During cough, the large arteries transmit increased intrathoracic pressure to the periphery, maintaining systemic pressure) -4,9,1~ However, although the gradient for peripheral blood flow during cough is increased, the gradient for coronary blood flow is decreased. 6,10, 12 Aortic pressure increases less than intrathoracic pressure 3, 10,11 or right atrial pressure, and thus coronary perfusion pressure (aortic minus right atrial pressure) decreases. Sustained increases in thoracic pressure (e.g., multiple coughs) lower aortic pressure and coronary blood flow velocity immediately and more so than single brief coughs in these normotensive subjects. The mechanisms of reduced coronary flow velocity reflected both the reduced aortic volume (and pressure) after cough and a possible reversal of left coronary flow due to mechanical compression of the coronary arteries 13 during cough. Transient reversal of coronary blood flow during ventricular systolic compression of the intramyocardial vessel has been demonstrated in patients with aortic stenosis 13 and in experimental animals with acute aortic insufficiency)4 In hypotensive patients, coughing may increase aortic flow.6 The sudden increase in intrathoracic pressure or external thoracic compression displaces peripheral blood from the central aorta and, in contrast to normotensive individuals, produces a gradient for opening of the aortic valve 6 but without a favorable left ventricular transmural gradient. The echocardiographic and hemodynamic demonstration of reduced coronary perfusion pressure by Little et al. 6 did not prove that coughing altered flow in coronary arteries. Reflex mechanisms during cough produce coronary vasodilatation in dogs) 5 However, the pulmonary inflation reflex in man is somewhat different from that in dogs) 6 In man, coronary enlargement during respiratory maneuvers has not been observed. 15 Effect of Valsalva maneuver ity.
and coronary
flow veloc-
The hemodynamics of the Valsalva maneuver in
Volume 119 Number 4
our patients are identical to those classical descriptions in which straining increases intrathoracic pressure abruptly, 17-19 The increased pressure is transmitted to the thoracic aorta with an increase in arterial blood pressure. Increased intrathoracic pressure reduces venous return, producing a decline in right and then in left ventricular end-diastolic volume, resulting in a fall of left ventricular stroke volume and a decline in mean arterial pressure and pulse pressure in phase III of the Valsalva maneuver. 19-21The fall in arterial pressure is accompanied by a reflex increase in heart rate and peripheral vasoconstriction. If autoregulation remains intact, coronary blood flow should decrease during the reduction of determinants of myocardial oxygen demand of the Valsalva maneuver. 23 In the current study, coronary flow velocity tended to decrease, but the decrease was not of the magnitude previously demonstrated. 15, 22 Coronary flow velocity was most significantly affected during phase III and remained relatively constant, adjusting to rapid changes in systemic pressure and heart rate. Benchimol et al., 22 measuring coronary ostial blood flow velocity, found similar changes of greater magnitude during the Valsalva maneuver. There was no difference in this response in patients with and without coronary artery disease. Wilson et a l ) 5 also found decreases in subselective coronary blood flow velocity in normal patients during the Valsalva maneuver. Changes in coronary flow velocity in 12 patients with normal coronary arteries and normal coronary flow reserve (4.8 _+ 0.20 units) during the Valsalva maneuver were different from the responses observed in dogs. Coronary blood flow velocity was reduced 0.67 _+ 0.9 times the resting value with increased coronary resistance (1.37 _+ 0.20 units, p = 0.05 versus control) during the Valsalva maneuver. No significant alterations in coronary blood flow velocity have been reported during deep breathing, the Muller maneuver, yawn, handgrip, and deep breathing during handgrip. 15 In patients, autonomic regulation of the coronary circulation does not appear to play a major role during the pulmonary inflation via reflex activation) 5 The patients in the current study differed from those of Wilson et al. 15 in that impaired coronary reserve was more prevalent and may play more of a role in attenuating the coronary flow response to the Valsalva maneuver. The current findings support previous results in which coronary flow decreases with increases in thoracic pressure. Whether reduced coronary reserve and some degree of coronary atherosclerosis alter coronary flow velocity responses remains under study. Limitations. Abnormalities that might affect the vasodilatory capacity of the arterial vasculature such as myocardial infarction, left ventricular hypertro-
Intrathoracic pressure and coronary flow
869
phy, left ventricular dysfunction, autonomic dysfunction, diabetes mellitus, orthostatic hypertension, Parkinson's disease, and renal dysfunction might produce different findings relative to coronary flow velocity changes in patients with normal coronary vasodilatory reserve. For this study, only one patient with left ventricular dysfunction was examined. Results are unchanged when this patient's responses are excluded. Doppler measurement of coronary blood flow velocity provides relative rather than absolute blood flow. Alterations in cross-sectional area of the artery during respiratory maneuvers or coughing might affect the relationship between flow and flow velocity. Utilizing quantitative angiography, Wilson et al. 15 demonstrated that coronary luminal area during the respiratory maneuvers described here in unchanged. Pharmacologic therapy may also influence the reflex responses. However, in similar patients with and without premedication, no differences were observed. 15Wilson et al) 5 noted a marked increase in coronary resistance during peak Valsalva maneuver, presumably due to sympathetic discharge and fall in coronary resistance during intracoronary papaverine, indicating that the coronary vasculature appeared to be capable of normal vasoconstriction and vasodilatation. Our study findings differ in that a majority of patients had a subnormal (<3.5 units) vasodilatory response. However, similar responses in peak and mean coronary flow velocity occurred for patients with and without coronary vasodilatory reserve >3.0 units. Also, similar coronary flow velocity responses were demonstrated for patients with and without coronary artery disease, 22 indicating little contribution of coronary disease to coronary flow responses to abrupt increases in intrathoracic pressure. Clinical significance. Abrupt increases in intrathoracic pressure (e.g., coughing) do not increase coronary blood flow velocity. Maintenance of arterial pressure for cerebral perfusion appears to be the major benefit during hypotensive conditions. The Valsalva maneuver tends to decrease coronary blood flow and is not significantly influenced by the decreased coronary perfusion pressure in patients with normal or impaired coronary reserve. The authors wish to thank the Mudd Cardiac Catheterization Team, and also Donna Sander for manuscript preparation. REFERENCES
1. Gensini GG. Coronary arteriography. In: Braunwald E, ed. Heart disease: A textbook of cardiovascular medicine. Philadelphia: WB Saunders Co, 1980:309. 2. Rudikoff MT, Maughan WL, Effron M, Freund P, Weisfeldt ML. Mechanisms of blood flow during cardiopulmonary resuscitation., Circulation 1980;61:345. 3. Rosborough JP, Hausknecht M, Criley JM, Garner D, Nie-
870
4. 5. 6.
7.
8.
9. 10. 11. 12.
Kern et at. mann JT. Cardiac output during ventricular fibrillation in the coughing dog [Abstract]. Am J Cardiol 1980;45:445. Sharpey-Schafer EP. Effects of coughing on intrathoracic pressure, arterial pressure and peripheral blood flow. J Physiol 1953;122:351. Pepine CJ. Coronary circulatory effects of increased intrathoracic pressure in intact dogs. Chest 1977;1:72. Little WC, Reeves RC, Coughlan C, Rogers EW. Effect of cough on coronary perfusion pressure: does coughing help clear the coronary arteries of angiographic contrast medium? Circulation 1982;65:604. Wilson RF, Laughlin DE, Ackell PH, Chilian WM, Holida MD, Hartley CJ, Armstrong ML, Marcus ML, White CW. Transluminal, subselective measurement of coronary artery blood flow velocity and vasodilator reserve in man. Circulation 1985;72:82. Kern MJ, Deligonul U, Vandormael M, Labovitz A, Gudipati CV, Gabliani G, Bodet J, Shah Y, Kennedy HL. Impaired coronary vasodilator reserve in the immediate postcoronary angioplasty period: analysis of coronary artery flow velocity indexes and regional cardial venous efflux. J Am Coll Cardiol 1989;13:860. Niemann JT, Garner D, Rosborougb J, Criley JM. The mechanism of blood flow in closed chest cardiopulmonary resuscitation [Abstract]. Circulation 1979;60(supp1 II):II-74. Hamilton WF, Woodbury RA, Harper HT. Arterial, cerebrospinal and venous pressures in man during cough and strain. Am J Physiol 1944;141:42. McIntosh HD, Estes EH, Warren JV. The mechanism of cough syncope. AM HEARTJ 1956;52:70. Cohen A, Gottdiener J, Wish M, Fletcher R. Limitations of cough in maintaining blood flow during asystole: assessment by two-dimensional and Doppler echocardiography. AMHEART J 1989;118:474.
April 1990 American Heart Journal
13. Carroll RJ, Falsetti HL. Retrograde coronary artery flow in aortic valve disease. Circulation 1976;54:494. 14. Folts JD, Rowe GG. Coronary and hemodynamic effects of temporary acute aortic insufficiency in intact anesthetized dogs. Circ Res 1974;35:238. 15. Wilson RF, Marcus ML, White CW. Pulmonary inflation reflex: its lack of physiological significance in coronary circulation of humans. Am J Physiol 1988;255(Heart Circ Physiol 24):H866. 16. Vatner SF, McRitchie RJ. Interaction of the chemoreflex and the pulmonary inflation reflex in the regulation of coronary circulation in conscious dogs. Circ Res 1975;37:664. 17. Hamilton WF, Woodbury RA, Harper HT Jr. Arterial cerebrospinal and venous pressures in man during cough and strain. Am J Physiol 1944;141:42. 18. Sharpey-Schafer EP. Effects of Valsalva's manoeuvre on the normal and failing circulation. Br Med J 1955;2:693. 19. Buda AJ, Pinsky MR, Ingels NB, Doughters GT, Stinson EB, Alderman EL. Effect of intrathoracic pressure on left ventricular performance. N Engl J Med 1979;301:453. 20. Brooker JZ, Alderman EL, Harrison DC. Alterations in left ventricular volumes induced by Valsalva manoeuvre. Br Heart J 1974;36:713. 21. Greenfield JC, Cox RL, Hernandex RR, Thomas C, Schoonmaker FW: Pressure-flow studies in man during the Valsalva maneuver with observations on the mechanical properties of the ascending aorta. Circulation 1967;35:653. 22. Benchimol A, Wang TF, Desser KB, Gartlan JL Jr. The Valsalva maneuver and coronary arterial blood flow velocity. Ann Intern Med 1972;77:357. 23. Pepine CJ, Wiener L. Effects of the Valsalva maneuver on myocardial ischemia in patients with coronary artery disease. Circulation 1979;59:1304.