SEPARATING DIASTOLIC DURATION AS CONFOUNDING VARIABLE IN THE HEMODYNAMIC EFFECT OF CORONARY VASODILATION

C. Kolyva*, B. Verhoeff**, J.J. Piek**, J.A.E. Spaan* and M. Siebes*

Departments of Medical Physics* and Cardiology**, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands

c.kolyva@amc.uva.nl

Abstract: The aim of this study was to quantify the effect of pharmacological vasodilation following coronary angioplasty and stenting on coronary microvascular resistance. Methods: Intracoronary a-receptor blockade was induced after stenting in 8 patients. Aortic pressure (Pa), distal coronary perfusion pressure (Pd) and blood flow velocity (v), and the electrocardiogram were recorded both at rest and during maximal hyperemia. Diastolic time fraction (DTF) and an index of coronary microvascular resistance during hyperemia (h-MRv = Pd/v) were determined per beat. Interacting hemodynamic effects induced by a-blockade were separated based on systolic/diastolic decomposition of h-MRv. Results: After a-blockade, Pa and Pd decreased by 8.0 ± 7.4 and 7.3 ± 7.2 mm Hg, respectively (p < 0.05). Irrespective of heart rate, DTF increased by 4.8 ± 2.6% (p < 0.005) via a reduction in systolic time by 5.6 ± 5.6% (p < 0.05). This increase in DTF accounted for a 16.2 ± 10.2% drop in average h-MRv, which, however, was masked by opposing effects of other hemodynamic changes. Conclusion: The method of analysis developed in this study successfully unraveled multiple opposing drug effects on the coronary microvascular resistance and identified an increase in DTF as a beneficial adjunct of a-receptor blockade.

Introduction

In the clinical setting of treating coronary arterial narrowings by balloon angioplasty and stenting, it is important to quantify and understand all factors involved in determining the resistance of the coronary microvascular bed [1]. Often, a drug is administered to test the reaction of the coronary vascular bed or to block the activity of vasoactive receptors in the vessel wall. In these kinds of studies, frequently only a single response to the drug is considered. However, many drugs interact not only with vascular smooth muscle, but they also affect the contraction pattern of the heart and thereby, the mechanical effects of the contracting heart muscle on coronary blood flow. Without a clear identification and separation of these confounding effects, no proper insight in the mechanistic action of a vasoactive drug can be obtained.

This study focuses on intracoronary hemodynamic measurements obtained during treatment with balloon angioplasty and/or stent placement. In this procedure, the cardiologist introduces a catheter into the femoral artery and advances it to the ostium of the right or left coronary artery. Via this hollow catheter a so-called guide wire with a diameter of 0.35 mm is advanced past the coronary lesion to be treated. In this study, a novel guide wire equipped with two sensors, one to measure blood flow velocity based on the Doppler principle and one for pressure measurement (Figure 1), was used.

After balloon dilatation of a coronary artery stenosis, coronary artery vasoconstriction frequently occurs as a result of the mechanical stretch of the arterial wall and/or distal embolization by debris that stimulates alpha-adrenergic pathways in the vessel wall, causing left ventricular dysfunction [2] and an impaired coronary flow reserve [3]. This adverse reaction can be abolished by subsequent administration of an a-receptor antagonist, leading to a recovery of the hemodynamic function by dilation of the constricted vessels [4].

Since perfusion of the heart muscle occurs predominantly during diastole, an increase in diastolic duration is beneficial, especially for patients with coronary artery disease. The aim of the present study was to find out whether administration of an a-receptor antagonist after PTCA acts not only by vasodilation of the resistance vessels, but also affects the myocardial perfusion by altering the mechanical characteristics of cardiac contraction. In order to do that, interacting effects of changes in heart rate, perfusion pressure, and vasomotor tone had to be separated from those of changes in the timing of cardiac contraction.

Materials and Methods

All data were obtained in the catheterization laboratory before and after percutaneous coronary intervention and analyzed off-line as described below. The study population consisted of eight patients (age 61 ± 7 yrs, 6 male) with coronary artery disease who were scheduled for elective coronary angioplasty and stenting and gave written informed consent.

Measurements: Aortic pressure (Pa) was measured via a 5F or 6F guiding catheter placed in the coronary ostium. Coronary perfusion pressure (Pd) and blood flow velocity (v) distal to the stenosis targeted for angioplasty were measured via a dual-sensor guide wire (Volcano Therapeutics Inc., Rancho Cordova, CA). These hemodynamic signals were recorded, together with the electrocardiogram (ECG), on a personal computer at a sampling rate of 120 Hz after 12-bit A/D conversion.

Protocol: All patients received intracoronary (i.c.) nitroglycerin (0.1 mg) at the beginning of the procedure in order to dilate the larger coronary vessels. Hemodynamic data were obtained at resting blood flow and during maximal hyperemia induced by a 20-40 µg i.c. bolus of adenosine. All signals were recorded before and after balloon angioplasty and stent placement, and 5-8 min following a 10 mg i.c. bolus of the selective a1-antagonist urapidil. This waiting period was necessary in order for urapidil to reach its maximum effect [3].

Data analysis: The digitized signals were analyzed using a custom-made program (written in Delphi version 6.0, Borland Software Corporation). First, the length of the cardiac cycle (T_RR) was determined as the time difference between two consecutive ECG R-peaks and heart rate (HR) was calculated as HR = 60/T_RR. Systole was defined as the period between the closing of the mitral valve, indicated by the ECG R-peak and the closing of the aortic valve, indicated by the dicrotic notch in the aortic pressure signal (Figure 2). In order to find the dicrotic notch, the local maximum of the first derivative of the aortic pressure signal after the peak of the ECG T-wave was located. By subtracting the systolic period from TRR we derived diastolic time (DT) and calculated the diastolic time fraction as:

DTF = DT/TRR (1)

The average change in DTF over all heart rates was obtained based on the mean difference between linear regression lines of DTF versus heart rate obtained before and after a-blockade.

Averages of aortic pressure, coronary perfusion pressure, and flow velocity were calculated per beat as well as for the systolic and diastolic period. From these data, a velocity-based index of microvascular resistance (MR_v) was then calculated per beat and for systole and diastole as MR_v = Pd/v [5]. Values representing resting conditions and hyperemia were obtained by averaging the corresponding per beat values. The hyperemic beats were defined as those (usually four or five) where microvascular resistance was minimum. Coronary flow velocity reserve (CFVR) was calculated as the ratio of hyperemic to resting flow velocity: CFVR = v_hyp/v_rest.

Hemodynamic effect of a-blockade due to alteration of cardiac contraction: The overall hemodynamic effect of a-blockade is reflected in a change in microvascular resistance during maximal hyperemia, when influences due to autoregulatory feedback can be excluded. Hyperemic MR_v (h-MR_v) is also dependent on perfusion pressure [6-7]. In order to separate a possible effect of a-blockade on the mechanical determinants of cardiac contraction (i.e., a change in DTF) from those of the direct dilatory action of urapidil and possible pressure-induced influences on h-MR_v, we used the following approach. For each cardiac cycle, the h-MR_v waveform was approximated by a square wave, representing average systolic and diastolic h-MR_v (Figure 3). With this, h-MR_v per beat could be expressed as:

h-MR_v = h-MR_{v sys} (1-DTF) + h-MR_{v dias} (DTF) (2)

The non-DTF-related changes in h-MR_v were then estimated by using the average values for systolic and diastolic h-MR_v after a-blockade and the values of DTF before a-blockade in Eq. 2. This is the value that h-MR_v would have if DTF had not changed after a-blockade. Therefore, the difference between this theoretical value and the h-MR_v that was actually measured after urapidil gives the change in h-MR_v that is due to DTF changes alone.

Statistics: Hemodynamic data obtained before and after a-blockade were compared using paired t-tests. Analysis of covariance (ANCOVA) was used to find whether there was a significant difference in the DTF-HR relation due to a-blockade. Statistical significance was assumed at p<0.05.

Results

Hemodynamic changes after a-blockade: During maximal hyperemia, aortic pressure after a-blockade decreased from 103.8 ± 12.4 mm Hg to 95.8 ± 9.6 mm Hg (p<0.05). Likewise, distal coronary pressure dropped from 98.0 ± 13.2 mm Hg to 90.7 ± 9.0 mm Hg (p<0.05). These changes were similar at resting conditions. No other hemodynamic parameters (heart rate, flow velocity, h-MR_v, or CFVR) changed significantly.

DTF changes after a-blockade: In all patients, there was a significant inverse relation between DTF and heart rate. The DTF-HR relation shifted upwards in 7 out of 8 patients after alpha-blockade, with a mean increase in DTF of 4.79 ± 2.62% (p<0.005) during maximal hyperemia for all heart rates. This increase was due to a decrease in systolic duration by 5.63 ± 5.55% (p<0.05). In one patient, a-blockade had no effect on DTF.

Effect of changes in DTF on h-MR_v: The respective DTF-related and non-DTF-related changes in h-MR_v are shown in Figure 4. While the overall change in h-MR_v after a-blockade was non-significant, the increase in DTF was responsible for a decrease in h-MR_v by 16.16 ± 10.19% (p<0.05). This decrease was, however, counteracted by other factors that tended to increase h-MR_v, such as a decrease in coronary perfusion pressure.

Discussion

With the approach described above, we were able to separate the influence of several interacting mechanisms that affect coronary hyperemic microvascular resistance in opposite directions after pharmacological dilation of the coronary microvessels due to a-receptor blockade.

Methodological considerations: An index of microvascular resistance was calculated based on flow velocity. Provided that the diameter of the vessel at the site of measurement remains constant, changes in h-MRv will be related to changes in flow in the same way as to changes in velocity. Administration of nitroglycerin was therefore necessary in order to keep the large epicardial arteries maximally dilated. Administration of adenosine, which targets the small coronary arteries and arterioles, on top of nitroglycerin ensured that the whole coronary vascular bed was maximally dilated.

For the calculation of DTF it is necessary to be able to measure accurately the duration of diastole. Ideally, phonocardiographic recordings or left ventricular pressure measurements would be necessary for this purpose, but they were not available in our experiments. It was however always possible to locate accurately the R-peaks of the ECG waveform and the dicrotic notch of the aortic pressure signal and calculate in a reproducible way the duration of systole and diastole. As long as the data are always analyzed in the same way, the calculated changes in DTF do not depend on the way systole and diastole are determined.

Coronary perfusion pressure and flow velocity should be preferably measured at the same position and not 3 cm apart. A time difference between the two signals would not be important for the calculation of the per beat averages, but it would be critical for the calculation of the systolic and diastolic averages. However, the combination of the delays due to the processing algorithm of the Doppler device and due to the processing of the pressure signal, led to simultaneous pressure and flow velocity measurements, despite the separation of the two transducers.

Significance of the study: The analysis of DTF is particularly important for the perfusion of the heart muscle close to the left ventricle, since this subendocardial layer is critically dependent on the duration of diastole [8].

During hyperemia, coronary flow is directly related to perfusion pressure since the resistance vessels are maximally dilated. Because of the distensibility of the vessels, coronary h-MR_v under these conditions is pressure-dependent and it varies inversely with perfusion pressure [6-7].

It is well known that DTF is inversely related to heart rate and our findings are consistent with those from previous studies [9-10]. It has also been shown that for a given heart rate, DTF increases in a nonlinear fashion with decreasing coronary perfusion pressure [11]. However, this effect is only relevant at perfusion pressures lower than 75 mm Hg, which was not the case in our patient group after stenting.

Our results provide evidence that a-blockade after coronary interventional treatment causes an increase in DTF, irrespective of the prevailing heart rate. This increase in DTF induces a decrease in h-MR_v, but that effect is concealed by the concurrent adverse changes in non-DTF-related factors, such as a drop in coronary perfusion pressure, which tends to increase h-MR_v. No hemodynamic effect of the pharmacological intervention could therefore be seen in any of the time-averaged parameters that are commonly used in clinical practice. Only by separately evaluating the systolic and diastolic components of the per beat averages could we reveal the beneficial hemodynamic action of this drug.

Conclusion

The method of analysis developed in this study may be very useful in unraveling multiple drug effects on the coronary circulation by other endogenous and exogenous stimuli.

Alpha-receptor blockade after PCI exerts a beneficial effect on hyperemic h-MR_v not only via adrenergic pathways, but also by prolongation of the duration of diastole irrespective of HR.

Acknowledgment

This study was supported in part by a grant from the Netherlands Heart Foundation (grant no. 2000.090).

References

1. L'ABBATE A., SAMBUCETI G., HAUNSO S. and SCHNEIDER-EICKE J. (1999): 'Methods for Evaluating Coronary Microvasculature in Humans', Eur. Heart J., 20, pp. 1300-1313
2. GREGORINI L., MARCO J., PALOMBO C., KOZAKOVA M., ANGUISSOLA G.B., CASSAGNEAU B., BERNIES M., DISTANTE A., MARCO I., FAJADET J. and ZANCHETTI A. (1998): 'Postischemic Left Ventricular Disfunction Is Abolished by Alpha-Adrenergic Blocking Agents', J. Am. Coll. Cardiol., 31(5), pp. 992-1001
3. GREGORINI L., MARCO J., FARAH B., BERNIES M., PALOMBO C., KOZAKOVA M., BOSSI I.M., CASSAGNEAU B., FAJADET J., DI MARIO C., ALBIERO R., CUGNO M., GROSSI A. and HEUSCH G. (2002): 'Effects of Selective a₁- and a₂- Adrenergic Blockade on Coronary Flow Reserve After Coronary Stenting', Circulation, 106, pp. 2901-2907
4. GREGORINI L., FAJADET J., ROBERT G., CASSAGNEAU B., BERNIS M. and MARCO J. (1994): 'Coronary Vasoconstriction After Percutaneous Transluminal Coronary Angioplasty Is Attenuated by Antiadrenergic Agents', Circulation, 90(2), pp. 895-907
5. SIEBES M., VERHOEFF B., MEUWISSEN M., DE WINTER R., SPAAN J.A.E. and PIEK J.J. (2004): 'Single-Wire Pressure and Flow Velocity Measurement to Quantify Coronary Stenosis Hemodynamics and Effects of Percuataneous Interventions', Circulation, 109, pp. 756-762
6. HANLEY F.L., MESSINA L.M., GRATTAN M.T. and HOFFMAN J.I.E. (1984): 'The Effect of Coronary Inflow Pressure on Coronary Vascular Resistance in the Isolated Dog Heart', Circ. Res., 54(6), pp. 760-772
7. CHAMULEAU S.A.J., SIEBES M., MEUWISSEN M., KOCH K.T., SPAAN J.A.E. and PIEK J.J. (2003): 'Association Between Coronary Lesion Severity and Distal Microvascular Resistance in Patients With Coronary Artery Disease', Am. J. Physiol., 285, pp. H2194-H2200
8. BACHE R.J. and COBB F.R. (1977): 'Effect of Maximal Coronary Vasodilation on Transmural Myocardial Perfusion During Tachycardia in the Awake Dog', Circ. Res., 41(5), pp. 648-653
9. BOUDOULAS H., LEWIS R.P., RITTGERS S.E., LEIER C.V. and VASKO J.S. (1979): 'Increased Diastolic Time: A Possible Important Factor in the Beneficial Effect of Propranolol in Patients with Coronary Artery Disease', J.Cardiovasc. Pharmacol., 1(5), pp. 503-513
10. FERRO G., GIUNTA A., MAIONE S., CARELLA G., GENOVESE A. and CHIARIELLO M. (1984): 'Diastolic Time During Exercise in Normal Subjects and in Patients with Coronary Artery Disease', Cardiology, 71, pp. 266-272
11. MERKUS D., KAJIYA F., VINK H., VERGROESEN I., DANKELMAN J., GOTO M. and SPAAN J.A.E. (1999): 'Prolonged Diastolic Time Fraction Protects Myocardial Perfusion When Coronary Blood Flow Is Reduced', Circulation, 100, pp. 75-81