Changes in systolic time intervals—a non-invasive marker for the haemodynamic effects of sumatriptan

Abstract

Aims

This study assessed the use of systolic time intervals (STI) as a potential non-invasive marker of the haemodynamic effects of sumatriptan, a 5HT₁ receptor agonist.

Methods

Twenty-six patients undergoing diagnostic cardiac catheterization participated. STIs were derived from haemodynamic pressure tracings at baseline, following placebo injection and following either subcutaneous (n = 18) or intravenous injection (n = 8) of sumatriptan.

Results

Sumatriptan (i.v. or s.c.) was associated with significant increases in mean arterial pressure (95% C.I. 9,14mmHg, P = 0.0001), total electromechanical systole (95% C.I.8,36ms, P<0.0001), pre-ejection period (95%C.I. 8,21ms, P = 0.0001) and left ventricular ejection time (95% C.I. 2,12ms, P = 0.004).

Conclusion

STI responses were consistent with sumatriptan-induced changes in afterload. In summary, the measurement of STIs is a potential non-invasive method of investigating the influence of serotonergic compounds on the cardiovascular system.

Introduction

Sumatriptan is a selective 5HT₁ agonist effective in the treatment of acute migraine [1, 2]. Reversal of cerebral arterial dilatation is thought to be the underlying mechanism for its efficacy in such situations [3]. Initial animal studies suggested that sumatriptan caused a selective vasoconstriction in the cranial circulation [4–6]. In humans, however, isolated case reports of chest pain with electrocardiographic changes raised the possibility of extracranial vasoconstrictive effects. Sumatriptan has since been shown in vivo to cause vasopressor responses in the systemic, pulmonary and coronary circulations when given intravenously [7] or subcutaneously [8, 9]. This suggests that 5HT₁ receptors exist in vascular beds distinct from the cranial circulation in humans. The therapeutic implications of this merit further investigation, both with sumatriptan and with other similar compounds. Previous observations used invasive techniques to assess the cardiovascular effects. These limit the potential for studies in patients and the identification of non-invasive physiological markers of vasoactive activity would be helpful to perform serial studies. The measurement of systolic time intervals (STI) represents a recognised sensitive, reproducible and repeatable method of investigating the cardioactive effects of drugs [10]. The aim of this study was therefore to calculate the systolic time intervals from pressure tracings in our previous invasive studies [7–9] and to determine whether STI may be used as a non invasive marker of the haemodynamic effects of sumatriptan.

Methods

Twenty-six patients (14M, 12F, age 49 years [s.d. 10]) previously studied during diagnostic cardiac catheterisation had pressure tracings suitable for analysis. Eight patients had previously participated in an intravenous protocol [7] and 18 had received subcutaneous sumatriptan [8, 9]. All vasoactive therapy other than sublingual GTN was discontinued 24 h before the study. Standard exclusion criteria were used to exclude women of childbearing potential, and patients with unstable clinical substrates including myocardial infarction within 3 months, unstable angina, cardiac arrhythmias and hypertension (diastolic BP >95 mmHg). Patients found to have coronary artery stenosis of >50% during diagnostic angiography were not entered.

This study was approved by the Ethics Committee of the Western Infirmary. Each patient was issued with an appropriate information sheet and written informed consent was obtained.

Left ventricular angiography and selective coronary arteriography were performed using the conventional Judkins technique via an arterial sheath placed in the right femoral artery. After the diagnostic procedure, the pigtail catheter (7F) was retained in the aorta and hard copy pressure tracings obtained throughout. Left ventricular pressure tracings were also obtained after crossing the aortic valve.

The systolic time intervals were measured from a simultaneous recorded electrocardiogram and aortic pulse wave tracings (Figure 1). This allowed the calculation of total electromechanical systole (EMS) from the onset of the QRS to the incisural notch on the arterial tracing. The left ventricular ejection time (LVET) was measured from the upstroke of the arterial tracing until the trough of the incisural notch. The pre-ejection period (PEP) was derived by subtracting the LVET from the total electromechanical systole. This represents the interval from the onset of electrical depolarisation until the onset of mechanical ejection. The ratio of the PEP over LVET was calculated (PEP/LVET). All STI values were then corrected for heart rate.

Figure 1.

Simultaneous tracing of electrocardiogram, phonocardiogram, left ventricular and aortic pressure tracings. LVET = left ventricular ejection time, PEP = pre-ejection period, QS₂ = total electromechanical systole.

Haemodynamic measurements were obtained at 10 min intervals until patient stability was confirmed (within 10%). The trial procedure is shown in Table 1. Following a placebo subcutaneous injection (18 patients) or intravenous placebo infusion (8 patients), serial measurements were obtained at 10 min and 20 min post injection. A subcutaneous injection of 6 mg sumatriptan (18 patients) or 10 min intravenous sumatriptan infusion to a maximum dose of 48 μg kg⁻¹ (8 patients) was then given. Haemodynamic measurements were recorded at 10 min intervals for 40 min. Aortic systolic, mean and diastolic pressure, heart rate and systolic time intervals were calculated for each time point.

Table 1.

Study protocol.

Data analysis

For each individual, a baseline value was obtained from two averaged observations. As any response occurred within 10 min and persisted for the 40 min study period, the average of the recordings at 10, 20, 30 and 40 min after sumatriptan injection was used to summarise the response to sumatriptan.

Statistics

Post-placebo and post-sumatriptan results were compared with baseline measurements using a paired t-test/confidence interval as the data appeared normally distributed.

Results

Results are shown in Figure 2 and in Table 2. Figures given in parentheses represent the 95% confidence intervals for the mean change from placebo values. Results were similar regardless of the route of drug administration.

Figure 2.

Mean (95% CI) increases in systemic arterial pressures and systolic time intervals after sumatriptan injection.

Table 2.

Arterial pressure and systolic time interval changes following placebo and sumatriptan. Baseline values are expressed as mean (s.d.). Statistical comparisons made were baseline versus placebo and placebo versus sumatriptan.

Electrocardiography

There were no changes in ECG morphology as assessed from the hard copies of six lead ECGs taken at 10 min intervals throughout the study.

Heart rate

Baseline heart rate was 68 (s.d. 14) beats min⁻¹. No significant change occurred after placebo injection. Following sumatriptan, heart rate decreased from the placebo value by 2 beats min⁻¹ (−4, 0 P = 0.05).

Systemic arterial pressures

Mean baseline values for systolic arterial pressure (SAP), diastolic arterial pressure (DAP) and mean arterial pressure (MAP) were 124, 73 and 94mmHg, respectively. There was no significant change between baseline and placebo values. After sumatriptan, systolic arterial pressure rose by 17 mmHg (12,22, P<0.0001). Diastolic arterial pressure rose by 6 mmHg (4,8, P<0.0001). Mean arterial pressure increased by 11 mmHg (8,14, P<0.0001).

Systolic time intervals

There were no significant differences between baseline and placebo values. Mean baseline (heart rate corrected) values for the systolic time intervals were; total electromechanical systole (EMS) 585ms, pre-ejection period (PEP) 163ms and left ventricular ejection time (LVET) 423 ms. Mean baseline PEP/LVET ratio was 0.39. After sumatriptan administration, EMS rose significantly by 22 ms (8,36, P<0.0001), PEP by 15 ms (8.21, P = 0.0001) and LVET by 7 ms (2, 12, P = 0.004). PEP/LVET ratio rose after sumatriptan by 0.03 (0.01, 0.04, P = 0.002).

Discussion

The present study assessed the measurement of systolic time intervals as a potential non-invasive marker for the effects of sumatriptan on the cardiovascular system. Conventionally, STIs are measured from simultaneous, fast recording, (i.e. paper speed >100 mm s⁻¹) of the electrocardiogram, phonocardiogram and carotid arterial pulsation. In this study STIs were calculated using high quality, fast paper speed arterial tracings and ECG recordings. A phonocardiogram was not recorded as STI measurements were based on invasive pressure recordings. The onset of the second heart sound (A₂) was therefore not recorded in this set up and so total electromechanical systole cannot be denoted by the conventional nomenclature (QS₂) but for purposes of this discussion will be referred to as EMS. The advantage of employing the sharp inscription of the incisural notch of the aortic pressure tracing as a marker of A₂ in this protocol is that the difficulty in identifying the variable first high frequency components of A₂ are not encountered.

The absolute duration of the systolic intervals in this study is longer than in those protocols utilising phonocardiograms as there is an inherent conduction delay in fluid filled catheters which will lead to a prolongation of the pre-ejection period and total electromechanical systole. The PEP consists of two sub intervals; electromechanical delay and isovolumic contraction time (ICT). In the absence of conduction disturbances (QRS<90 ms) the ICT and PEP are closely correlated [10]. In this study no sequential changes in ECG morphology were noted and it is therefore assumed that the change in PEP reflects prolongation of ICT.

This study demonstrates that sumatriptan significantly prolongs EMS, PEP and LVET with a consequent rise in the PEP/LVET ratio. We assume that the observed changes in the STIs are drug related rather than procedural effects. The observed rise in PEP and PEP:LVET in response to sumatriptan may have occurred secondary to a negative inotropic effect or due to an acute rise in afterload. Analysis of the overall change in all STIs allows the precise mode of action to be established. Negative inotropes would be expected to decrease LVET without significant effect on the EMS, a pattern observed by Boudoulas et al. when studying the negative inotropic effects of lignocaine [11]. Acute increases in afterload however prolong the EMS and LVET as well as the PEP and PEP:LVET and this pattern resembles closely the change in STIs following sumatriptan injection. Also against a negative inotropic action is the lack of a previous invasive study to show a significant change in the peak rate of rise of LV pressure (P/t) following sumatriptan [9]. These STI changes are therefore consistent with a sumatriptan-induced increase in afterload as witnessed by the rise in systemic arterial pressures.

Increases in afterload induced by intravenous injection of methoxamine are, like sumatriptan, accompanied by a lengthening of the PEP, QS₂ and LVET [12]. Shaver et al. maintained heart rate constant with atrial pacing during methoxamine infusion, thereby preventing reflex bradycardia and noted an approximate 1 ms increase in LVET for each mmHg rise in systolic arterial pressure [12]. Angiotensin infusion, an agent with almost exclusive changes in afterload however, causes increases in QS₂ and PEP, but with a reduction in LVET [13]. Increases in PEP:LVET are seen with acute increases in afterload and in response to drugs with negative inotropic effects [14].

The effect of sumatriptan on pre-load is less clear. Increases in left ventricular end-diastolic pressure (LVEDP) and pulmonary artery wedge pressure (PAWP) may be accounted for by increases in pre-load and afterload [9]. A prolongation of EMS following sumatriptan is also in keeping with an increase in pre-load but the PEP would normally shorten as pre-load increases. It is therefore assumed that, although sumatriptan may augment pre-load, the over-riding effect seems to be an increase in afterload.

Invasive haemodynamic studies of serotonergic compounds have obvious limitations and therefore the use of systolic time intervals as a non-invasive marker of haemodynamics has appeal. For example, the duration of action of serotonergic compounds could be assessed utilising the STI changes as physiological markers of the pharmacological activity. In this study the STI changes were evident within 10 min, in keeping with a sumatriptan t_max of 5-20 min [15] and a 15 min response rate of 74% in cluster headache [16]. The duration of the response in this study was maintained to at least 40 min, consistent with sumatriptan’s 1 h anti-migraine response rate of 77% [17].

In conclusion, it has been shown that sumatriptan induces changes in STIs consistent with its effect on afterload, and that the measurement of systolic time intervals may represent an accurate, non-invasive method of investigating the influence of serotonergic compounds on the cardiovascular system. This requires confirmation in a protocol that measures STI using a phonocardiogram, carotid arterial waveforms and the electrocardiogram.