We analyzed the Valsalva maneuver (VM) using a photoplethysmographic technique in 24 normal subjects. We studied the effects of test duration (range 5 to 20 seconds) and magnitude of expiratory pressure (EP, range 20 to 50 mmHg) on maximal (%A) and time integrated (int) changes in heart rate (HR), and mean (MAP) and pulse (PP) pressures during phases II and IV. At all EPs, there was correlation (P < 0.01) between o/oA MAP in both late phase I I and phase IV and duration of the VM. The Valsalva ratio (VR) correlated with test duration only at high EPs. All responses, except the fall in MAP in early phase II and bradycardia in phase IV, correlated with EPs. The VR correlate not only with HR responses in phases I I and IV but also with MAP and PP overshoot in phase IV. Correlations between maximal AP and HR changes in either phase II and IV were poor. Our study indicates that the photoplethysmographic technique allows a more rational interpretation of the VM in the clinical setting. The VR cannot be adequately interpreted without corncornitant monitoring of arterial pressure during the VM. Key words: forced expiration heart rate arterial pressure noninvasive monitoring MUSCLE & NERVE 14~1165-1172 1991
USE OF THE PHOTOPLETHYSMOGRAPHIC TECHNIQUE TO ANALYZE THE VALSALVA MANEUVER IN NORMAL MAN EDUARDO E. BENARROCH, MD, TONETTE L. OPFER-GEHRKING, and PHILLIP A. LOW, MD
Cardiovascular tests of autonomic function have been used extensively in clinical laboratories. The Valsalva maneuver (VM) consists of an abrupt transient voluntary elevation of intrathoracic and intra-abdominal pressure provoked by forced expiration against The hemodynamic changes during the VM have been well documented using invasive technique^,^.^"^'^^ and can be affected by multiple variables, including the magnitude and duration of the expiratory effort.6 In normal subjects, the responses of the VM can be divided in 4 phases’: phase I, consisting of an evanescent rise in arterial pressure (AP) and
From the Autonomic Reflex Laboratory, Mayo Foundation, Rochester, Minnesota. Acknowledgments: Supported in part by grants from NINCDS (NS14304, NS22302, and NS72302). MDA. and by the Mogg Funds. Dr. Low is the recipient of a Jacob Javits Neuroscience Investigator Award from NINCDS. Address reprint requests to Eduardo E Benarroch. MD. Autonomic Reflex Laboratory, Mayo Foundation, Rochester, MN 55905. Accepted for publication October 16, 1990. CCC 0148-639)(/91/01201165-08 $04.00 0 1991 John Wiley & Sons, Inc.
Finapres Technique in Valsalva Maneuver
decrease in heart rate (HR) immediately after the onset of straining; phase 11, characterized by a fall, and later recovery of AP and an acceleration of HR during the period of straining; phase 111, consisting of a sudden brief reduction of AP and increase in HR immediately following the release of straining; and phase IV, characterized by an elevation of AP above control levels (overshoot) and slowing of the HR. Pharmacological studies have indicated that, whereas phases I and 111 are mainly due to mechanical factors, the changes in MAP and HR during phases I1 and IV depend on complex mechanoreceptor reflex inputs and involve both cardiovagal and sympathetic effector mechanisms.” In the clinical setting, the VM has been commonly used to calculate the Valsalva ratio (VR), defined as the ratio of the longest R-R interval after the maneuver to the shortest R-R interval during the m a n e ~ v e r . The ’ ~ VR has been assumed to be a test of cardiovagal f ~ n c t i o n . ’ ~Although ”~ the VR is reproducible in normal subjects, and has proven useful to detect autonomic dysfunction, 1*2 it only provides a partial and indirect assessment of the functional integrity of the reflex mecha-
MUSCLE & NERVE
December 1991
1165
nisms involved in the cardiovascular responses during the VM. Intra-arterial recording of AP has provided important additional information about the hemodynamic changes during the VM,12721 but its use has been restricted to research laboratories due to the invasiveness of the procedure, which limits its application in the clinical setting. Recent evidence indicates that the photoplethysmographic (Finapres) technique for continuous, noninvasive AP recording reliably reflects AP values obtained with intra-arterial recordings, 16,17320,24325including the changes in AP during the VM.' In the present study, we further explored the value of the Finapres technique in the assessment of autonomic responses during the VM, which could be applied clinically. We investigated the influence of the test duration and expiratory pressure on AP and HR responses during the VM, using a range of test conditions commonly used in clinical laboratories, and assessed the correlation between HR and AP responses during different phases of the VM. Our results indicate that the Finapres technique provides information that is not only comparable to that obtained from invasive procedures, but that could also allow a more rational interpretation of the responses to the VM in the clinical setting. MATERlALS AND METHODS
Twenty-four healthy subjects (14 males, 10 females), aged 20 to 47 years (36.4 2 ) , were studied. They were allowed to rest for at least 10 minutes before the procedure. Room temperature was kept at 23°C. Heart rate (HR) was continuously recorded via ECG leads connected to a cardiac monitor (Physio Control Lifepak 8). Systolic (SAP), diastolic AP (DAP), and mean arterial pressures (MAP) were continuously monitored using the photoplethysmographic technique (Finapres).17,20 An inflatable cuff, containing the transducer of an infrared transmission plethysmograph, was wrapped around the middle phalanx of a finger and, by means of an active servonull mechanism, a counterpressure was generated whose values faithfully reproduce the blood volume pulsations under the cuff.
Subjects and Procedures.
The subjects were studied in the supine position and performed the expiratory effort on command by blowing through a mouthpiece connected to a mercury manometer. The
signal for commencing forced expiration was given at the end of a deep inspiration. A small leak in the system prevented undetected closure of the glottis, thus ensuring that the expiratory pressure was transmitted to the alveoli. The expiratory pressure was monitored throughout the maneuver using a pressure transducer. The subjects were instructed to breathe normally after the expiratory effort, and were allowed to rest for at least 3 minutes after stabilization of AP and HR before repeating the maneuver. PROTOCOL
The study was approved by the Institutional Review Board, and informed consent was obtained from the volunteers. We sought to separately determine the effects of test duration and effort on AP and HR responses during VM. We chose the range of test conditions most commonly used clinically, and included low pressures and short duration to evaluate a weak effort as might be expected from ill patients. T o study the effects of test duration, 20 subjects (1 1 M, 9 F) were instructed to sustain the expiratory effort for periods of 5, 10, 15, or 20 seconds. The VM was performed first at low EP (20 mmHg), and then at high EP (50 mmHg). A resting period of 2 to 3 minutes after stabilization of AP and HR was allowed between tests. T h e effects of the magnitude of EP were examined separately in 10 subjects (6 M, 4 F) who were instructed to keep an expiratory effort of 20, 30, 40, or 50 mmHg for a period of 15 seconds. Stabilization between tests was allowed, as described above. Analog data from the various devices were fed to an analog to a digital converter (MICRO 6809) then input into an AST 286 personal computer for display, storage, and data manipulation. The computer console provided a continuous display of HR, MAP, SAP, DAP, and expiratory effort in real time. Computer hardware and software was developed by Irvin Zimmerman (MSEE), Section of Engineering. The program was written in C language and permitted operator selectable, second-by-second readout of each of the parameters, the time integrals, and their first derivatives.
Storage and Analysis of the Data.
Valsalva Maneuver.
1166
Finapres Technique in Valsalva Maneuver
PARAMETERS ANALYZED
The 4 phases of the VM8 were clearly identified in every subject. VR, and maximal percent (%A)
MUSCLE & NERVE
December 1991
changes ([maximal - basal/basal] x 100) in HR and mean (MAP), systolic (SAP), diastolic (DAP), and pulse (PP) arterial pressures were calculated from direct readings on the computer screen. T h e MAP was automatically derived from the pulse wave. It corresponds approximately to DAP
+
1.-(
The time integral (int) of changes in MAP in phase I1 (int MAP 11) was obtained by integrating the area from the peak of MAP in phase I to the nadir of MAP in phase 11. The peak MAP in phase I, rather than resting MAP, was chosen to calculate int MAP I1 because, in many subjects, MAP did not fall below basal during phase II. T h e time integral of changes in HR during phase I1 (int HR 11), MAP during phase IV (int MAP IV), and HR during phase IV (int HR IV) were calculated by integrating the area above or below baseline, respectively. Recovery time was calculated from the peak of MAP and HR change during phase IV to the return of HR and MAP to baseline. STATISTICAL ANALYSIS
Data were analyzed by multiple comparison analysis of variance, unpaired t test, or least squares reI
Ill
4 II 4 IV
gression analysis, as appropriate. Results are expressed as mean + SEM. P < 0.05 was considered significant. RESULTS Cardiovascular Responses to VM Monitored With the Finapres Technique. All subjects showed the typi-
cal AP and HR changes during the 4 phases of the Valsalva maneuver. An example is shown in Figure 1. Regardless of the effort and duration of VM, there were no significant differences between females and males in any of the parameters analyzed. There was no correlation between any of the responses to VM and age. Therefore, data from all subjects were combined. Effects of Test Duration on Responses to VM.
Twenty subjects performed VM, first against a resistance of 20 mmHg for 5, 10, 15, or 20 seconds; and then against a resistance of 50 mmHg for 5, 10, 15, or 20 seconds. Resting HR (70 2 2 beats/ min) and MAP (81.8 5 1.8 mmHg) were not significantly different at the beginning of each test. In general, there was low correlation between the HR and AP changes and the duration of the VM (Table 1). T h e maximal increase in HR during phase I1 of the VM and the VR correlated with the duration of the VM, but only when performed
Table 1. Correlation of heart rate and arterial pressure responses with duration of the Valsalva maneuver. Expiratory pressure (mmHg)
L
I
15 sec
FIGURE 1. Diagram representing the changes in heart rate (HR), mean arterial pressure (MAP), and pulse pressure (BP) during the 4 phases (I, II, 111, and IV) of the Valsalva maneuver in a 34-year-old female. The lower bar represents the duration of the maneuver performed at an expiratory pressure of 40 mmHg.
Finapres Technique in Valsalva Maneuver
Parameter
20 (r)
50 (r)
Valsalva ratio
0.09
0.43'
% A MAP I
0.10 0.07
0.12 0.41* -0.07
Yo A HR II % A MAP I1 (e) % A MAP II (I) Int A HR I1 Int A MAP II Yo A HR IV % A MAP IV Int A HR IV Int A MAP IV Recovery HR Recoverv MAP
-0.10
0.43t 0.47' -0.37t -0.08
0.48t 0.7V -0.59t -0.12
0.25$
0.33$ 0.37$
0.26$
0.26$
0.27t
0.30t 0.34t
0.22*
0.35t
Number of cases: 20; test duration: 5, 10, 15, 20 seconds; % A: percent change; lnt: time integral; MAP. mean arterial pressure; HR- heart fate; I , I / , lV. phases of the Valsalva maneuver; (e): early; 0): late; r: correlation coefficient; NS: nonsignificant; 'P < 0.0001; t P i0.001; #P < 0.02.
MUSCLE & NERVE
December 1997
1167
at high EPs. On the other hand, the maximal changes in MAP in late phase I1 and phase IV significantly correlated with duration of the VM as both low (20 mmHg) and high (50 mmHg) expiratory pressures (Fig. 2).
Table 2. Correlation of heart rate and arterial pressure responses with expiratory effort during the Valsalva maneuver. Parameters
Effects of Expiratory Effort on Responses to Valsalva Maneuver. Ten subjects performed the VM
against a resistance of 20, 30, 40, or 50 mmHg for 15 seconds. In general, the changes of AP and HR during VM showed stronger correlation with the magnitude of EPs (Table 2) than with the duration of VM (Table 1). Most parameters showed statistically significant correlation with EPs (Fig. 3 ) . Correlation Among Parameters of the Valsalva Maneuver. The correlations between changes of HR
Valsalva ratio % A MAP I %AHRI % A HR II % A MAP I I (e) % A MAP I I (I) %APPII
0.63' 0.71* -0.18 0.62' -0.02 0.62* -0.41'
Int HR II Int MAP II
0.82' -0.40t
% A HR IV % A MAP IV % A DAP IV Int HR IV Int MAP IV
-0.17 0.58' 0.34$ -0.41 t 0.53*
Recovery HR Recovery MAP
and AP in different phases of the VM are summarized in Table 3. Valsalva Ratio. The VR showed stronger correlation with the magnitude of changes in HR during phase I1 than during phase IV. There was also a significant correlation between the VR and the magnitude of AP overshoot in phase IV (Fig. 4). In each subject, there was a strong correlation (r = 0.95) between the VR and the increase in MAP during phase I. In the group as a whole, this correlation was also significant (Table 3 ) . Phase ZZ. In general, there was poor correlation between maximal HR and AP responses during phase 11. The only significant (but low) corre-
0.58* 0.63'
Number of cases: 70; expiratory effort: 20, 30, 40, 50 mmHg; test duration: 15 seconds; % A: percent change; lnt: time integral; MAP: mean arterial pressure; DAP: diastolic arterial pressure; PP: pulse pressure; HR: heart rate; I, I / , IV: phases of the Valsalva maneuver; (e): early; (I). late; r = correlation coefficient. 'P < 0.0007; t P -=z 0.001; #P < 0.02.
lations were those between the maximal increase in HR and decrease in PP, and between the timeintegrated HR and MAP responses. Phase IV. There was no correlation between
MAP in late phase It
Valsalva ratio 25 0
EP 50 m m Hg
0 EP 20 m m Hg
20
a:
>
15
10 0
5
10
15
20
Test duration, sec
25
0
5
10
15
20
25
Test duration. sec
FIGURE 2. Effects of duration of the Valsalva maneuver on the Valsalva ratio (VR) (left), and the percent increase (change %) in mean arterial pressure (MAP) in late phase II (right). For each test duration, the maneuver was performed at EP (expiratory pressure) of 20 or 50 mmHg. Values are expressed as mean c SEM. Number of subjects: 20.
1168
Finapres Technique in Valsalva Maneuver
MUSCLE & NERVE
December 1991
Phase II
Phase IV 80
0 HR
0 HR
- A MAP (I) 60
-
PP
40
8 ai
-
A
F'20
m
6
0
-20
-40
1 0 2 0 3 0 4 0 5 0 6 0
Expiratory pressure, mm Hg
10
20
30
40
50
60
Expiratory pressure, mm Hg
FIGURE 3. Effect of the magnitude of expiratory pressure during the Valsalva maneuver on changes (change %) of heart rate (HR), mean arterial pressure (MAP), and pulse pressure (PP) during early (e) and late (I) phase II (left) and during phase IV (right). Values are expressed as mean 2 SEM. Number of subjects: 10. Test duration: 15 seconds.
maximal decrease in HR and maximal overshoot of MAP, SAP, or PP during phase IV. Heart rate responses showed only mild, although significant, correlation with DAP overshoot. On the other hand, there was correlation between time-integrated MAP and HR responses and, even higher correlation between the time of recovery of MAP and HR following the VM (Fig. 5). Correlation Between AP Changes in Phase 11 and Phase IV. There was a direct correlation between the maximal increase in MAP in late phase I1 and in phase IV, and an inverse correlation between the changes in DAP in phase IV and in PP during phase 11.
Effects of Test Duration and Expiratory Effort on HR and AP Responses During the VM. Both the mag-
nitude of the expiratory pressure and the duration of the straining during the VM may affect
Table 3. Correlation between heart rate and arterial pressure responses during the Valsalva maneuver. Parameter
Correlation with:
Valsalva ratio
% A HR II % A HR IV
0.74* 0.59'
Yo A MAP IV % A DAP IV Yo A PP IV
0.60' 0.56t 0.48'
DISCUSSION
T h e value of the Finapres technique for continuing monitoring of AP changes during the VM has been demonstrated in previous s t ~ d i e s . ~ . ' Our ~.*~ results indicate that this technique, in combination with continuous heart rate recording, can be applied to obtain a more meaningful interpretation of the VM in the clinical setting, without the use of invasive procedures. The independent influence of the test duration and expiratory pressures on the AP and HR responses, and the correlation between these changes in the different phases of VM, may provide indication on the main physiological mechanisms involved, especially when compared with data obtained using invasive intra-arterial recordings.
Finapres Technique in Valsalva Maneuver
% A MAP I
% A HR II
0.52*
Yo A MAP II (e) % A MAP II (I) Yo A PP II
Int HR II
Int MAP II
-0.08 -0.10 -0.48t -0.487
% A HR IV
Int HR IV
% A MAP IV % A DAP IV Int MAP IV
-0.22 -0.44$ -0.58*
HR recovery time Yo A MAP IV Yo A DAP IV
MAP recovery time % A MAP II (I) Yo A PAP II
0.96' 0.50' -0.50"
Number of cases: 70;% A. percent change; lnt: time integral; MAP: mean arterial pressure; DAP: diastolic pressure; PP: pulse pressure; HR: heart rate; I . I / , IV: phases of the Valsalva maneuver; (ej: early; (I): late; r : correlation coefficient. "P < 0.0001; t P < 0.001; t P < 0.01.
MUSCLE & NERVE
December 1991
1169
0.0 0
1
I
I
1
I
25
50
15
100
125
h
150
% HR (11)
-
0.0 -60
30
40
-20
-10
0
a%HR(n/)
0.0
1 1 1 1 1 1 1 1 1 1 ) 0
102030405060108090100 A
% MAP (IV)
finding that, at low EPs the reflex tachycardia in phase I1 does not correlate with test duration, may also be explained by the conditions of our testing. In the supine position, there is a large intrathoracic blood volume which buffers the reduced venous return during phase Il.21225At high EPs, this thoracic vascular buffer mechanism is insufficient and the magnitude of reflex tachycardia in phase 11, and thus the VR, correlates with the duration of the VM. The finding that the maximal increase in MAP in late phase I1 and in phase IV correlated with the duration of the VM is consistent with the idea that these responses mainly reflect the activity of long latency sympathetic mechanisms. * Sympathetic blockade exaggerates the fall of MAP in phase 11 and attenuates A P overshoot in phase IV.I2 It has been suggested that 30 seconds is probably the minimum time necessary to reach stability of sympathetic responses.' However, the finding that microneurographically recorded muscle sympathetic nerve activity increases within 3 to 4 seconds of VM and is, followed, within 4 seconds, by a cessation of a decline of MAP in phase 11,5 suggests that adequate vasoconstrictor responses
'*"
0 -250 -500
FIGURE 4. Correlationsbetween the Valsalva ratio and the maximal changes (change YO)of heart rate (HR) in phase I1 (upper panel), HR in phase IV (middle panel), and mean arterial pressure (MAP) in phase IV (lower panel).
-750 -1,OOO -1.250
.___
-1.m
0
250
500
750
1,Ooo 1.250 1.500
.
MAP time integral, mm Hg sec
mechanical and reflex hemodynamic responses.6"2 This may be due, at least in part, to the differences in the threshold of peripheral mechanoreceptors and in the time constant of vagal and sympathetically mediated responses. Test durations of 10 to 20 seconds and expiratory efforts of 20 to 50 mmHg are most commonly used clinically122*6 and were, therefore, explored in this study. Effects of Test Duration. Our results show that, at low EPs, the magnitude of tachycardia during phase I1 is independent of the duration of the VM. This is consistent with the concept, based on pharmacological studies, that this response is primarily mediated by inhibition of vagal mechan i s m ~ which , ~ ~ have a short time constant." T h e
'',"
1170
Finapres Technique in Valsalva Maneuver
150 125 100
75 50 25 0
0
25
50
75
100
125
150
Time of MAP recovery, sec FIGURE 5. Upper panel: Correlation between the tirne-integrated changes in heart rate (HR) and mean arterial pressure (MAP) during phase IV of the Valsalva maneuver. Lower panel: Correlation between the time of recovery of heart rate (HR) and mean arterial pressure (MAP) after release of expiratory effort.
MUSCLE 8, NERVE
December 1991
can be achieved earlier than 30 seconds. This is of practical importance, because it is difficult for many subjects, especially those with autonomic failure, to sustain high EPs for long periods. Our results indicate that a test duration of 10 seconds is effective, and 15 seconds is a practical optimum, and this may be sufficient to assess sympathetically mediated responses in the clinical setting. Effects of Expiratory Pressure. Our results are consistent with those obtained using invasive techniques,“ and show that the magnitude of most HR and AP responses during phases I1 and IV correlate with the magnitude of EP employed during the VM. I n early phase 11, the decrease in PP, but not MAP, correlated with the EP. This suggests that arterial baroreflexes, which respond mainly to changes in pulsatile pressure,“ play an increasingly important role at higher EPs, due to a progressive decrease in PP (presumably due to a decrease in cardiac output), relative to MAP. T h e finding that the increase in MAP in late phase I1 reflects sympathetic vasoconstriction is supported by its positive correlation with the magnitude of EP, which is consistent with the evidence of a progressive increase in total peripheral resistance with increasing EPs. l 2
Correlations Between HR and MAP Responses Dur. ing VM. T h e VR showed a better correlation with
the magnitude of HR response in phase I1 than in phase IV. This indicates that, if tachycardia in phase I1 is present, the VR may be “normal,” even in the absence of significant bradycardia in phase IV. This may occur in patients with cardiovagal impairment but intact cardiac sympathetic innervation.15 That the VR correlates with AP overshoot in phase IV, indicates that the absence of bradycardia in phase IV may not only be due to vagal dysfunction, but also to inability to increase MAP and thus stimulate the baroreflex. In summary, the interpretation of the VR as a test of “cardiovagal function” in the absence of simultaneous recording of AP, may be an oversimplification. T h e HR responses during the VM may be critically affected by both (a) the magnitude of decrease in venous return, which depends on the position of the subject and on the pooling and buffering effect of thoracic vessels; and by (b) the changes in AP during phase I1 and IV. T h e complex mechanical and reflex cardiovascular mechanisms, as well as the respiratory changes7 during the VM, make it difficult to use
Finapres Technique in Valsalva Maneuver
the correlation between maximal changes in AP and HR as an isolated index of baroreceptor function. We were unable to find a correlation between the maximum AP and HR responses during the VM. Correlation between AP and HR changes during the VM could be observed, however, by analyzing the temporal profile of the responses (i.e., time-integral) and, particularly, the time of recovery of MAP and HR during phase IV. Clinical Implications. Our results indicate that the VM performed with EPs of 40 mmHg for 15 seconds, as commonly used clinically, adequately stimulates cardiopulmonary and baroreceptor input, and allows assessment of cardiovagal and sympathetic functions. Whereas test duration predominantly influences indices of sympathetic function, the magnitude of expiratory effort significantly affects both sympathetic and cardiovagal responses. Interpretation of heart rate data in the absence of concomitant measurement of AP may be misleading. Changes in the heart rate, and therefore the VR, should be evaluated taking into account the changes in AP. Exaggerated fall of MAP in phase I1 due to vasoconstrictor failure may produce an exaggerated tachycardia, and thus increase of the VR; whereas lack of overshoot of MAP in phase IV may abolish bradycardia in phase IV, and thus decrease the VR. The magnitude of decrease in MAP in phase I1 and overshoot in phase IV have been considered indices of vasomotor f u n ~ t i o n . ~Based ” ~ on the present data, we suggest that the magnitude of increase of MAP in late phase I1 may be a more useful index of sympathetic vasoconstriction. T h e fall of MAP in early phase I1 is affected by the magnitude of HR responses,” and the overshoot of AP in phase IV may be dependent on the cardiac output. This is supported by our recent studies with the sequential blockade of sympathetic vasoconstrictor, sympathetic cardiac, and vagal inputs.lg T h e use of the Finapres technique has clear clinical application, and this is reflected in our finding that patients with mild sympathetic dysfunction may have absent increase in MAP in late phase I1 and phase IV, before showing either a decrease in MAP in early phase I1 or clinical orthostatic hypotension. Combination of noninvasive AP monitoring with microneurographic recording of peripheral sympathetic activity may provide an objective electrodiagnostic measurement of the integrity of peripheral vasoconstrictor innervation in patients with autonomic failure.
’*
MUSCLE & NERVE
December 1991
1171
REFERENCES 1. Baldwa VS, Ewing DJ: Heart rate response to Valsalva manoeuvre. Reproducibility in normals, and relation to variation in resting heart rate in diabetics. Br Heart J 1977;39:641-644. 2. Bennett T, Farquhar IK, Hosking DJ, Hampton JR: An assessment of methods of estimating autonomic nervous system control of the heart in patients with diabetes mellitus. Diabetes 1978;27:1174-1177. 3. Booth RW, Ryan JM, Mellett HC, Swiss E, Neth E: Hemodynamic changes associated with the Valsalva maneuver in normal men and women. J La6 Clin Med 1962;59:275285. 4. Corbett JL: Some aspects of the autonomic nervous system in normal and abnormal man. DPhil Thesis, University of Oxford, 1969. 5. Delius W, Hagbarth KE, Hongell A, Wallin BG: Manoeuvres affecting sympathetic outflow in human muscle nerves. Acta Physiol Scand 1972;84:82-94. 6. Eckberg DL: Parasympathetic cardiovascular control in human disease: a critical review of methods and results. A m J Physiol 1980;239:H581- H593. 7. Eckberg DL, Orshan CR: Respiratory and baroreceptor reflex interactions in man. J Clin Invest 1977;59:780-785. 8. Hamilton WF, Woodbury RA, Harper HT Jr: Physiological relationships between intrathoracic, intraspinal and arterial pressures. J A M A 1936;107:53-56. 9. Imholz BPM, van Montfrans GA, Settels JJ, van der Hoeven GMA, Karemaker JM, Wieling w : Continuous noninvasive blood pressure monitoring: reliability of Finapres device during the Valsalva manoeuvre. Cardiov Res 1988;22:390- 397. 10. Johnson RH, Spalding JMK: Disorders of the Autonomic Nervous System, Oxford, Blackwell, 1974. 11. Korner PI: Integrative neural cardiovascular control. Physiol Rev 1971;51:312-367. 12. Korner PI, Tonkin AM, Uther JB: Reflex and mechanical circulatory effects of graded Valsalva maneuvers in normal man. J Appl Physiol 1976;40:434-440. 13. Leon DF, Shaver JA, Leonard JJ: Reflex heart rate control in man. A m Heart J 1976;80:727-739. 14. Levin AB: A simple test of cardiac function based upon the heart rate changes induced by the Valsalva manoeuvre. A m J Cardiol 1966;18:90-99. 15. Low PA, Walsh JC, Huang CY, McLeod JG: The sympa-
1172
Finapres Technique in Valsalva Maneuver
15. Low PA, Walsh JC, Huang CY, McLeod JG: The sympathetic nervous system in diabetic neuropathy: a clinical and pathological study. Brain 1975;98:341- 356. 16. Parati G , Casadei R, Gropelli A, Di Rienzo M, Mancia G: Comparison of finger and intra-arterial blood pressure monitoring at rest and during laboratory testing. Hypertension 1989;13:647-655. 17. Penaz J: Photoelectric measurement of blood pressure, volume and flow in the finger, in Albert R, Vogt WS, Helbig W (eds): Digest of the International Conference on Medicine and Biologzcal Engineering. Conference Committee of the Xth International Conference on Medicine and Biological Engineering, Dresden, 1973, p 104. 18. Sandroni P, Benarroch EE, Low PA: The detection of sympathetic adrenergic vasomotor failure by analysis of BP components during Valsalva maneuver. Neurology 1990;4O(suppl 1):386. 19. Sandroni P, Benarroch EE, Low PA: Pharmacological dissections of the components of the Valsalva maneuver in adrenergic failure. J Appl Physiol (in press). 20. Settels JJ, Wesseling KH: Finuprest Nonznvasive Finger Arterial Pressure Waveform Regstration. Psychophysiolop~of Cardiovascular Control. New York, Plenum, 1985, pp 267-283. 2 1. Sharpey-Schafer EP: Effects of Valsalva’s maneouvre in normal and failing circulation. Br Med J 1955;1:693-695. 22. ten Harkel ADJ, van Lieshout EJ, van Lieshout W, Wieling W: The assessment of cardiovascular reflexes, influence of posture and period of preceding rest. J Appl Physiol 1989 (in press). 23. Thames MD, Kontos HA: Mechanisms of baroreceptor-induced changes in heart rate. A m J Physiol 1970;218:251256. 24. Van Egmond J, Hasenbos M, Crul JF: Invasive v. non-invasive measurement of arterial pressure. Comparison of two automatic methods and simultaneously measured direct intra-arterial pressure. B r J Anesth 1985;57:434-444. 25. Yamakoshi K, Shimazu H, Togawa T: Indirect measurement of instantaneous arterial blood pressure in the human finger by the vascular unloading technique. IEEE Trans Biomed Eng 1980;27: 150- 155. 26. Zema MJ, Restivo B, Sos T, Sniderman KW, Skline S: Left ventricular dysfunction-bedside Valsalva maneuvre. Br Heart J 1980;44:560-569.
MUSCLE & NERVE
December 1991
USE OF THE PHOTOPLETHYSMOGRAPHIC TECHNIQUE TO ANALYZE THE VALSALVA MANEUVER IN NORMAL MAN EDUARDO E. BENARROCH, MD, TONETTE L. OPFER-GEHRKING, and PHILLIP A. LOW, MD
Cardiovascular tests of autonomic function have been used extensively in clinical laboratories. The Valsalva maneuver (VM) consists of an abrupt transient voluntary elevation of intrathoracic and intra-abdominal pressure provoked by forced expiration against The hemodynamic changes during the VM have been well documented using invasive technique^,^.^"^'^^ and can be affected by multiple variables, including the magnitude and duration of the expiratory effort.6 In normal subjects, the responses of the VM can be divided in 4 phases’: phase I, consisting of an evanescent rise in arterial pressure (AP) and
From the Autonomic Reflex Laboratory, Mayo Foundation, Rochester, Minnesota. Acknowledgments: Supported in part by grants from NINCDS (NS14304, NS22302, and NS72302). MDA. and by the Mogg Funds. Dr. Low is the recipient of a Jacob Javits Neuroscience Investigator Award from NINCDS. Address reprint requests to Eduardo E Benarroch. MD. Autonomic Reflex Laboratory, Mayo Foundation, Rochester, MN 55905. Accepted for publication October 16, 1990. CCC 0148-639)(/91/01201165-08 $04.00 0 1991 John Wiley & Sons, Inc.
Finapres Technique in Valsalva Maneuver
decrease in heart rate (HR) immediately after the onset of straining; phase 11, characterized by a fall, and later recovery of AP and an acceleration of HR during the period of straining; phase 111, consisting of a sudden brief reduction of AP and increase in HR immediately following the release of straining; and phase IV, characterized by an elevation of AP above control levels (overshoot) and slowing of the HR. Pharmacological studies have indicated that, whereas phases I and 111 are mainly due to mechanical factors, the changes in MAP and HR during phases I1 and IV depend on complex mechanoreceptor reflex inputs and involve both cardiovagal and sympathetic effector mechanisms.” In the clinical setting, the VM has been commonly used to calculate the Valsalva ratio (VR), defined as the ratio of the longest R-R interval after the maneuver to the shortest R-R interval during the m a n e ~ v e r . The ’ ~ VR has been assumed to be a test of cardiovagal f ~ n c t i o n . ’ ~Although ”~ the VR is reproducible in normal subjects, and has proven useful to detect autonomic dysfunction, 1*2 it only provides a partial and indirect assessment of the functional integrity of the reflex mecha-
MUSCLE & NERVE
December 1991
1165
nisms involved in the cardiovascular responses during the VM. Intra-arterial recording of AP has provided important additional information about the hemodynamic changes during the VM,12721 but its use has been restricted to research laboratories due to the invasiveness of the procedure, which limits its application in the clinical setting. Recent evidence indicates that the photoplethysmographic (Finapres) technique for continuous, noninvasive AP recording reliably reflects AP values obtained with intra-arterial recordings, 16,17320,24325including the changes in AP during the VM.' In the present study, we further explored the value of the Finapres technique in the assessment of autonomic responses during the VM, which could be applied clinically. We investigated the influence of the test duration and expiratory pressure on AP and HR responses during the VM, using a range of test conditions commonly used in clinical laboratories, and assessed the correlation between HR and AP responses during different phases of the VM. Our results indicate that the Finapres technique provides information that is not only comparable to that obtained from invasive procedures, but that could also allow a more rational interpretation of the responses to the VM in the clinical setting. MATERlALS AND METHODS
Twenty-four healthy subjects (14 males, 10 females), aged 20 to 47 years (36.4 2 ) , were studied. They were allowed to rest for at least 10 minutes before the procedure. Room temperature was kept at 23°C. Heart rate (HR) was continuously recorded via ECG leads connected to a cardiac monitor (Physio Control Lifepak 8). Systolic (SAP), diastolic AP (DAP), and mean arterial pressures (MAP) were continuously monitored using the photoplethysmographic technique (Finapres).17,20 An inflatable cuff, containing the transducer of an infrared transmission plethysmograph, was wrapped around the middle phalanx of a finger and, by means of an active servonull mechanism, a counterpressure was generated whose values faithfully reproduce the blood volume pulsations under the cuff.
Subjects and Procedures.
The subjects were studied in the supine position and performed the expiratory effort on command by blowing through a mouthpiece connected to a mercury manometer. The
signal for commencing forced expiration was given at the end of a deep inspiration. A small leak in the system prevented undetected closure of the glottis, thus ensuring that the expiratory pressure was transmitted to the alveoli. The expiratory pressure was monitored throughout the maneuver using a pressure transducer. The subjects were instructed to breathe normally after the expiratory effort, and were allowed to rest for at least 3 minutes after stabilization of AP and HR before repeating the maneuver. PROTOCOL
The study was approved by the Institutional Review Board, and informed consent was obtained from the volunteers. We sought to separately determine the effects of test duration and effort on AP and HR responses during VM. We chose the range of test conditions most commonly used clinically, and included low pressures and short duration to evaluate a weak effort as might be expected from ill patients. T o study the effects of test duration, 20 subjects (1 1 M, 9 F) were instructed to sustain the expiratory effort for periods of 5, 10, 15, or 20 seconds. The VM was performed first at low EP (20 mmHg), and then at high EP (50 mmHg). A resting period of 2 to 3 minutes after stabilization of AP and HR was allowed between tests. T h e effects of the magnitude of EP were examined separately in 10 subjects (6 M, 4 F) who were instructed to keep an expiratory effort of 20, 30, 40, or 50 mmHg for a period of 15 seconds. Stabilization between tests was allowed, as described above. Analog data from the various devices were fed to an analog to a digital converter (MICRO 6809) then input into an AST 286 personal computer for display, storage, and data manipulation. The computer console provided a continuous display of HR, MAP, SAP, DAP, and expiratory effort in real time. Computer hardware and software was developed by Irvin Zimmerman (MSEE), Section of Engineering. The program was written in C language and permitted operator selectable, second-by-second readout of each of the parameters, the time integrals, and their first derivatives.
Storage and Analysis of the Data.
Valsalva Maneuver.
1166
Finapres Technique in Valsalva Maneuver
PARAMETERS ANALYZED
The 4 phases of the VM8 were clearly identified in every subject. VR, and maximal percent (%A)
MUSCLE & NERVE
December 1991
changes ([maximal - basal/basal] x 100) in HR and mean (MAP), systolic (SAP), diastolic (DAP), and pulse (PP) arterial pressures were calculated from direct readings on the computer screen. T h e MAP was automatically derived from the pulse wave. It corresponds approximately to DAP
+
1.-(
The time integral (int) of changes in MAP in phase I1 (int MAP 11) was obtained by integrating the area from the peak of MAP in phase I to the nadir of MAP in phase 11. The peak MAP in phase I, rather than resting MAP, was chosen to calculate int MAP I1 because, in many subjects, MAP did not fall below basal during phase II. T h e time integral of changes in HR during phase I1 (int HR 11), MAP during phase IV (int MAP IV), and HR during phase IV (int HR IV) were calculated by integrating the area above or below baseline, respectively. Recovery time was calculated from the peak of MAP and HR change during phase IV to the return of HR and MAP to baseline. STATISTICAL ANALYSIS
Data were analyzed by multiple comparison analysis of variance, unpaired t test, or least squares reI
Ill
4 II 4 IV
gression analysis, as appropriate. Results are expressed as mean + SEM. P < 0.05 was considered significant. RESULTS Cardiovascular Responses to VM Monitored With the Finapres Technique. All subjects showed the typi-
cal AP and HR changes during the 4 phases of the Valsalva maneuver. An example is shown in Figure 1. Regardless of the effort and duration of VM, there were no significant differences between females and males in any of the parameters analyzed. There was no correlation between any of the responses to VM and age. Therefore, data from all subjects were combined. Effects of Test Duration on Responses to VM.
Twenty subjects performed VM, first against a resistance of 20 mmHg for 5, 10, 15, or 20 seconds; and then against a resistance of 50 mmHg for 5, 10, 15, or 20 seconds. Resting HR (70 2 2 beats/ min) and MAP (81.8 5 1.8 mmHg) were not significantly different at the beginning of each test. In general, there was low correlation between the HR and AP changes and the duration of the VM (Table 1). T h e maximal increase in HR during phase I1 of the VM and the VR correlated with the duration of the VM, but only when performed
Table 1. Correlation of heart rate and arterial pressure responses with duration of the Valsalva maneuver. Expiratory pressure (mmHg)
L
I
15 sec
FIGURE 1. Diagram representing the changes in heart rate (HR), mean arterial pressure (MAP), and pulse pressure (BP) during the 4 phases (I, II, 111, and IV) of the Valsalva maneuver in a 34-year-old female. The lower bar represents the duration of the maneuver performed at an expiratory pressure of 40 mmHg.
Finapres Technique in Valsalva Maneuver
Parameter
20 (r)
50 (r)
Valsalva ratio
0.09
0.43'
% A MAP I
0.10 0.07
0.12 0.41* -0.07
Yo A HR II % A MAP I1 (e) % A MAP II (I) Int A HR I1 Int A MAP II Yo A HR IV % A MAP IV Int A HR IV Int A MAP IV Recovery HR Recoverv MAP
-0.10
0.43t 0.47' -0.37t -0.08
0.48t 0.7V -0.59t -0.12
0.25$
0.33$ 0.37$
0.26$
0.26$
0.27t
0.30t 0.34t
0.22*
0.35t
Number of cases: 20; test duration: 5, 10, 15, 20 seconds; % A: percent change; lnt: time integral; MAP. mean arterial pressure; HR- heart fate; I , I / , lV. phases of the Valsalva maneuver; (e): early; 0): late; r: correlation coefficient; NS: nonsignificant; 'P < 0.0001; t P i0.001; #P < 0.02.
MUSCLE & NERVE
December 1997
1167
at high EPs. On the other hand, the maximal changes in MAP in late phase I1 and phase IV significantly correlated with duration of the VM as both low (20 mmHg) and high (50 mmHg) expiratory pressures (Fig. 2).
Table 2. Correlation of heart rate and arterial pressure responses with expiratory effort during the Valsalva maneuver. Parameters
Effects of Expiratory Effort on Responses to Valsalva Maneuver. Ten subjects performed the VM
against a resistance of 20, 30, 40, or 50 mmHg for 15 seconds. In general, the changes of AP and HR during VM showed stronger correlation with the magnitude of EPs (Table 2) than with the duration of VM (Table 1). Most parameters showed statistically significant correlation with EPs (Fig. 3 ) . Correlation Among Parameters of the Valsalva Maneuver. The correlations between changes of HR
Valsalva ratio % A MAP I %AHRI % A HR II % A MAP I I (e) % A MAP I I (I) %APPII
0.63' 0.71* -0.18 0.62' -0.02 0.62* -0.41'
Int HR II Int MAP II
0.82' -0.40t
% A HR IV % A MAP IV % A DAP IV Int HR IV Int MAP IV
-0.17 0.58' 0.34$ -0.41 t 0.53*
Recovery HR Recovery MAP
and AP in different phases of the VM are summarized in Table 3. Valsalva Ratio. The VR showed stronger correlation with the magnitude of changes in HR during phase I1 than during phase IV. There was also a significant correlation between the VR and the magnitude of AP overshoot in phase IV (Fig. 4). In each subject, there was a strong correlation (r = 0.95) between the VR and the increase in MAP during phase I. In the group as a whole, this correlation was also significant (Table 3 ) . Phase ZZ. In general, there was poor correlation between maximal HR and AP responses during phase 11. The only significant (but low) corre-
0.58* 0.63'
Number of cases: 70; expiratory effort: 20, 30, 40, 50 mmHg; test duration: 15 seconds; % A: percent change; lnt: time integral; MAP: mean arterial pressure; DAP: diastolic arterial pressure; PP: pulse pressure; HR: heart rate; I, I / , IV: phases of the Valsalva maneuver; (e): early; (I). late; r = correlation coefficient. 'P < 0.0007; t P -=z 0.001; #P < 0.02.
lations were those between the maximal increase in HR and decrease in PP, and between the timeintegrated HR and MAP responses. Phase IV. There was no correlation between
MAP in late phase It
Valsalva ratio 25 0
EP 50 m m Hg
0 EP 20 m m Hg
20
a:
>
15
10 0
5
10
15
20
Test duration, sec
25
0
5
10
15
20
25
Test duration. sec
FIGURE 2. Effects of duration of the Valsalva maneuver on the Valsalva ratio (VR) (left), and the percent increase (change %) in mean arterial pressure (MAP) in late phase II (right). For each test duration, the maneuver was performed at EP (expiratory pressure) of 20 or 50 mmHg. Values are expressed as mean c SEM. Number of subjects: 20.
1168
Finapres Technique in Valsalva Maneuver
MUSCLE & NERVE
December 1991
Phase II
Phase IV 80
0 HR
0 HR
- A MAP (I) 60
-
PP
40
8 ai
-
A
F'20
m
6
0
-20
-40
1 0 2 0 3 0 4 0 5 0 6 0
Expiratory pressure, mm Hg
10
20
30
40
50
60
Expiratory pressure, mm Hg
FIGURE 3. Effect of the magnitude of expiratory pressure during the Valsalva maneuver on changes (change %) of heart rate (HR), mean arterial pressure (MAP), and pulse pressure (PP) during early (e) and late (I) phase II (left) and during phase IV (right). Values are expressed as mean 2 SEM. Number of subjects: 10. Test duration: 15 seconds.
maximal decrease in HR and maximal overshoot of MAP, SAP, or PP during phase IV. Heart rate responses showed only mild, although significant, correlation with DAP overshoot. On the other hand, there was correlation between time-integrated MAP and HR responses and, even higher correlation between the time of recovery of MAP and HR following the VM (Fig. 5). Correlation Between AP Changes in Phase 11 and Phase IV. There was a direct correlation between the maximal increase in MAP in late phase I1 and in phase IV, and an inverse correlation between the changes in DAP in phase IV and in PP during phase 11.
Effects of Test Duration and Expiratory Effort on HR and AP Responses During the VM. Both the mag-
nitude of the expiratory pressure and the duration of the straining during the VM may affect
Table 3. Correlation between heart rate and arterial pressure responses during the Valsalva maneuver. Parameter
Correlation with:
Valsalva ratio
% A HR II % A HR IV
0.74* 0.59'
Yo A MAP IV % A DAP IV Yo A PP IV
0.60' 0.56t 0.48'
DISCUSSION
T h e value of the Finapres technique for continuing monitoring of AP changes during the VM has been demonstrated in previous s t ~ d i e s . ~ . ' Our ~.*~ results indicate that this technique, in combination with continuous heart rate recording, can be applied to obtain a more meaningful interpretation of the VM in the clinical setting, without the use of invasive procedures. The independent influence of the test duration and expiratory pressures on the AP and HR responses, and the correlation between these changes in the different phases of VM, may provide indication on the main physiological mechanisms involved, especially when compared with data obtained using invasive intra-arterial recordings.
Finapres Technique in Valsalva Maneuver
% A MAP I
% A HR II
0.52*
Yo A MAP II (e) % A MAP II (I) Yo A PP II
Int HR II
Int MAP II
-0.08 -0.10 -0.48t -0.487
% A HR IV
Int HR IV
% A MAP IV % A DAP IV Int MAP IV
-0.22 -0.44$ -0.58*
HR recovery time Yo A MAP IV Yo A DAP IV
MAP recovery time % A MAP II (I) Yo A PAP II
0.96' 0.50' -0.50"
Number of cases: 70;% A. percent change; lnt: time integral; MAP: mean arterial pressure; DAP: diastolic pressure; PP: pulse pressure; HR: heart rate; I . I / , IV: phases of the Valsalva maneuver; (ej: early; (I): late; r : correlation coefficient. "P < 0.0001; t P < 0.001; t P < 0.01.
MUSCLE & NERVE
December 1991
1169
0.0 0
1
I
I
1
I
25
50
15
100
125
h
150
% HR (11)
-
0.0 -60
30
40
-20
-10
0
a%HR(n/)
0.0
1 1 1 1 1 1 1 1 1 1 ) 0
102030405060108090100 A
% MAP (IV)
finding that, at low EPs the reflex tachycardia in phase I1 does not correlate with test duration, may also be explained by the conditions of our testing. In the supine position, there is a large intrathoracic blood volume which buffers the reduced venous return during phase Il.21225At high EPs, this thoracic vascular buffer mechanism is insufficient and the magnitude of reflex tachycardia in phase 11, and thus the VR, correlates with the duration of the VM. The finding that the maximal increase in MAP in late phase I1 and in phase IV correlated with the duration of the VM is consistent with the idea that these responses mainly reflect the activity of long latency sympathetic mechanisms. * Sympathetic blockade exaggerates the fall of MAP in phase 11 and attenuates A P overshoot in phase IV.I2 It has been suggested that 30 seconds is probably the minimum time necessary to reach stability of sympathetic responses.' However, the finding that microneurographically recorded muscle sympathetic nerve activity increases within 3 to 4 seconds of VM and is, followed, within 4 seconds, by a cessation of a decline of MAP in phase 11,5 suggests that adequate vasoconstrictor responses
'*"
0 -250 -500
FIGURE 4. Correlationsbetween the Valsalva ratio and the maximal changes (change YO)of heart rate (HR) in phase I1 (upper panel), HR in phase IV (middle panel), and mean arterial pressure (MAP) in phase IV (lower panel).
-750 -1,OOO -1.250
.___
-1.m
0
250
500
750
1,Ooo 1.250 1.500
.
MAP time integral, mm Hg sec
mechanical and reflex hemodynamic responses.6"2 This may be due, at least in part, to the differences in the threshold of peripheral mechanoreceptors and in the time constant of vagal and sympathetically mediated responses. Test durations of 10 to 20 seconds and expiratory efforts of 20 to 50 mmHg are most commonly used clinically122*6 and were, therefore, explored in this study. Effects of Test Duration. Our results show that, at low EPs, the magnitude of tachycardia during phase I1 is independent of the duration of the VM. This is consistent with the concept, based on pharmacological studies, that this response is primarily mediated by inhibition of vagal mechan i s m ~ which , ~ ~ have a short time constant." T h e
'',"
1170
Finapres Technique in Valsalva Maneuver
150 125 100
75 50 25 0
0
25
50
75
100
125
150
Time of MAP recovery, sec FIGURE 5. Upper panel: Correlation between the tirne-integrated changes in heart rate (HR) and mean arterial pressure (MAP) during phase IV of the Valsalva maneuver. Lower panel: Correlation between the time of recovery of heart rate (HR) and mean arterial pressure (MAP) after release of expiratory effort.
MUSCLE 8, NERVE
December 1991
can be achieved earlier than 30 seconds. This is of practical importance, because it is difficult for many subjects, especially those with autonomic failure, to sustain high EPs for long periods. Our results indicate that a test duration of 10 seconds is effective, and 15 seconds is a practical optimum, and this may be sufficient to assess sympathetically mediated responses in the clinical setting. Effects of Expiratory Pressure. Our results are consistent with those obtained using invasive techniques,“ and show that the magnitude of most HR and AP responses during phases I1 and IV correlate with the magnitude of EP employed during the VM. I n early phase 11, the decrease in PP, but not MAP, correlated with the EP. This suggests that arterial baroreflexes, which respond mainly to changes in pulsatile pressure,“ play an increasingly important role at higher EPs, due to a progressive decrease in PP (presumably due to a decrease in cardiac output), relative to MAP. T h e finding that the increase in MAP in late phase I1 reflects sympathetic vasoconstriction is supported by its positive correlation with the magnitude of EP, which is consistent with the evidence of a progressive increase in total peripheral resistance with increasing EPs. l 2
Correlations Between HR and MAP Responses Dur. ing VM. T h e VR showed a better correlation with
the magnitude of HR response in phase I1 than in phase IV. This indicates that, if tachycardia in phase I1 is present, the VR may be “normal,” even in the absence of significant bradycardia in phase IV. This may occur in patients with cardiovagal impairment but intact cardiac sympathetic innervation.15 That the VR correlates with AP overshoot in phase IV, indicates that the absence of bradycardia in phase IV may not only be due to vagal dysfunction, but also to inability to increase MAP and thus stimulate the baroreflex. In summary, the interpretation of the VR as a test of “cardiovagal function” in the absence of simultaneous recording of AP, may be an oversimplification. T h e HR responses during the VM may be critically affected by both (a) the magnitude of decrease in venous return, which depends on the position of the subject and on the pooling and buffering effect of thoracic vessels; and by (b) the changes in AP during phase I1 and IV. T h e complex mechanical and reflex cardiovascular mechanisms, as well as the respiratory changes7 during the VM, make it difficult to use
Finapres Technique in Valsalva Maneuver
the correlation between maximal changes in AP and HR as an isolated index of baroreceptor function. We were unable to find a correlation between the maximum AP and HR responses during the VM. Correlation between AP and HR changes during the VM could be observed, however, by analyzing the temporal profile of the responses (i.e., time-integral) and, particularly, the time of recovery of MAP and HR during phase IV. Clinical Implications. Our results indicate that the VM performed with EPs of 40 mmHg for 15 seconds, as commonly used clinically, adequately stimulates cardiopulmonary and baroreceptor input, and allows assessment of cardiovagal and sympathetic functions. Whereas test duration predominantly influences indices of sympathetic function, the magnitude of expiratory effort significantly affects both sympathetic and cardiovagal responses. Interpretation of heart rate data in the absence of concomitant measurement of AP may be misleading. Changes in the heart rate, and therefore the VR, should be evaluated taking into account the changes in AP. Exaggerated fall of MAP in phase I1 due to vasoconstrictor failure may produce an exaggerated tachycardia, and thus increase of the VR; whereas lack of overshoot of MAP in phase IV may abolish bradycardia in phase IV, and thus decrease the VR. The magnitude of decrease in MAP in phase I1 and overshoot in phase IV have been considered indices of vasomotor f u n ~ t i o n . ~Based ” ~ on the present data, we suggest that the magnitude of increase of MAP in late phase I1 may be a more useful index of sympathetic vasoconstriction. T h e fall of MAP in early phase I1 is affected by the magnitude of HR responses,” and the overshoot of AP in phase IV may be dependent on the cardiac output. This is supported by our recent studies with the sequential blockade of sympathetic vasoconstrictor, sympathetic cardiac, and vagal inputs.lg T h e use of the Finapres technique has clear clinical application, and this is reflected in our finding that patients with mild sympathetic dysfunction may have absent increase in MAP in late phase I1 and phase IV, before showing either a decrease in MAP in early phase I1 or clinical orthostatic hypotension. Combination of noninvasive AP monitoring with microneurographic recording of peripheral sympathetic activity may provide an objective electrodiagnostic measurement of the integrity of peripheral vasoconstrictor innervation in patients with autonomic failure.
’*
MUSCLE & NERVE
December 1991
1171
REFERENCES 1. Baldwa VS, Ewing DJ: Heart rate response to Valsalva manoeuvre. Reproducibility in normals, and relation to variation in resting heart rate in diabetics. Br Heart J 1977;39:641-644. 2. Bennett T, Farquhar IK, Hosking DJ, Hampton JR: An assessment of methods of estimating autonomic nervous system control of the heart in patients with diabetes mellitus. Diabetes 1978;27:1174-1177. 3. Booth RW, Ryan JM, Mellett HC, Swiss E, Neth E: Hemodynamic changes associated with the Valsalva maneuver in normal men and women. J La6 Clin Med 1962;59:275285. 4. Corbett JL: Some aspects of the autonomic nervous system in normal and abnormal man. DPhil Thesis, University of Oxford, 1969. 5. Delius W, Hagbarth KE, Hongell A, Wallin BG: Manoeuvres affecting sympathetic outflow in human muscle nerves. Acta Physiol Scand 1972;84:82-94. 6. Eckberg DL: Parasympathetic cardiovascular control in human disease: a critical review of methods and results. A m J Physiol 1980;239:H581- H593. 7. Eckberg DL, Orshan CR: Respiratory and baroreceptor reflex interactions in man. J Clin Invest 1977;59:780-785. 8. Hamilton WF, Woodbury RA, Harper HT Jr: Physiological relationships between intrathoracic, intraspinal and arterial pressures. J A M A 1936;107:53-56. 9. Imholz BPM, van Montfrans GA, Settels JJ, van der Hoeven GMA, Karemaker JM, Wieling w : Continuous noninvasive blood pressure monitoring: reliability of Finapres device during the Valsalva manoeuvre. Cardiov Res 1988;22:390- 397. 10. Johnson RH, Spalding JMK: Disorders of the Autonomic Nervous System, Oxford, Blackwell, 1974. 11. Korner PI: Integrative neural cardiovascular control. Physiol Rev 1971;51:312-367. 12. Korner PI, Tonkin AM, Uther JB: Reflex and mechanical circulatory effects of graded Valsalva maneuvers in normal man. J Appl Physiol 1976;40:434-440. 13. Leon DF, Shaver JA, Leonard JJ: Reflex heart rate control in man. A m Heart J 1976;80:727-739. 14. Levin AB: A simple test of cardiac function based upon the heart rate changes induced by the Valsalva manoeuvre. A m J Cardiol 1966;18:90-99. 15. Low PA, Walsh JC, Huang CY, McLeod JG: The sympa-
1172
Finapres Technique in Valsalva Maneuver
15. Low PA, Walsh JC, Huang CY, McLeod JG: The sympathetic nervous system in diabetic neuropathy: a clinical and pathological study. Brain 1975;98:341- 356. 16. Parati G , Casadei R, Gropelli A, Di Rienzo M, Mancia G: Comparison of finger and intra-arterial blood pressure monitoring at rest and during laboratory testing. Hypertension 1989;13:647-655. 17. Penaz J: Photoelectric measurement of blood pressure, volume and flow in the finger, in Albert R, Vogt WS, Helbig W (eds): Digest of the International Conference on Medicine and Biologzcal Engineering. Conference Committee of the Xth International Conference on Medicine and Biological Engineering, Dresden, 1973, p 104. 18. Sandroni P, Benarroch EE, Low PA: The detection of sympathetic adrenergic vasomotor failure by analysis of BP components during Valsalva maneuver. Neurology 1990;4O(suppl 1):386. 19. Sandroni P, Benarroch EE, Low PA: Pharmacological dissections of the components of the Valsalva maneuver in adrenergic failure. J Appl Physiol (in press). 20. Settels JJ, Wesseling KH: Finuprest Nonznvasive Finger Arterial Pressure Waveform Regstration. Psychophysiolop~of Cardiovascular Control. New York, Plenum, 1985, pp 267-283. 2 1. Sharpey-Schafer EP: Effects of Valsalva’s maneouvre in normal and failing circulation. Br Med J 1955;1:693-695. 22. ten Harkel ADJ, van Lieshout EJ, van Lieshout W, Wieling W: The assessment of cardiovascular reflexes, influence of posture and period of preceding rest. J Appl Physiol 1989 (in press). 23. Thames MD, Kontos HA: Mechanisms of baroreceptor-induced changes in heart rate. A m J Physiol 1970;218:251256. 24. Van Egmond J, Hasenbos M, Crul JF: Invasive v. non-invasive measurement of arterial pressure. Comparison of two automatic methods and simultaneously measured direct intra-arterial pressure. B r J Anesth 1985;57:434-444. 25. Yamakoshi K, Shimazu H, Togawa T: Indirect measurement of instantaneous arterial blood pressure in the human finger by the vascular unloading technique. IEEE Trans Biomed Eng 1980;27: 150- 155. 26. Zema MJ, Restivo B, Sos T, Sniderman KW, Skline S: Left ventricular dysfunction-bedside Valsalva maneuvre. Br Heart J 1980;44:560-569.
MUSCLE & NERVE
December 1991
