Effect of Etilefrine in Preventing Syncopal Recurrence in Patients With Vasovagal Syncope
A Double-Blind, Randomized, Placebo-Controlled Trial
Background—Etilefrine is an α-agonist agent with a potent vasoconstrictor effect, which is potentially useful in preventing vasovagal syncope by reducing venous pooling and/or by counteracting reflex arteriolar vasodilatation. The present multicenter, randomized, placebo-controlled study was designed to evaluate the efficacy of this drug for the long-term management of patients with recurrent vasovagal syncope.
Methods and Results—In the 20 participating centers, 126 patients with recurrent vasovagal syncope (at least 3 episodes in the last 2 years) and a positive baseline head-up tilt response were randomly assigned to placebo (63 patients) or etilefrine at a dosage of 75 mg/d (63 patients) and were followed up for 1 year or until syncope recurred. The primary end-point of the study was the first recurrence of syncope. There were no differences between the 2 study groups in the patients’ baseline characteristics. During follow-up, the group treated with etilefrine had a similar incidence of first syncopal recurrence to that of placebo group both in the intention-to-treat analysis (24% versus 24%) and in on- treatment analysis (26% versus 24%). Moreover, the median time to the first syncopal recurrence did not significantly differ between the 2 study groups (106 days in the etilefrine arm and 112 days in the placebo arm).
Conclusions—Oral etilefrine is not superior to placebo in preventing spontaneous episodes of vasovagal syncope. Randomized controlled studies are essential to assess the real usefulness of any proposed therapy for patients with vasovagal syncope.
The exact pathophysiological mechanism of vasovagal syncope is still unclear; nonetheless, a key component of the process seems to be an increase in venous pooling leading to a reduction in venous return with resultant paradoxical activation of ventricular mechanoreceptors, reflex sympathetic withdrawal, and arteriolar vasodilatation.1 2 3 4 5 6 α-sympathetic agents are thought to prevent vasovagal syncope by 2 mechanisms.7 8 First, by increasing venous tone, these drugs may reduce venous pooling and thus possibly avoid the paradoxical activation of ventricular mechanoreceptors. Second, by increasing arteriolar tone, they may counterbalance the profound reflex vasodilatation and the hypotension that are ultimately responsible for the syncope. Etilefrine is an α-agonist agent with a potent vasoconstrictor effect that was first used in 1990 for treatment of vasovagal syncope. In a small, initial, open study,9 the drug was shown to be effective in preventing both tilt-induced and spontaneous episodes of vasovagal syncope. However, subsequent studies performed with this and other sympathetic agents yielded conflicting results.8 10 11 12 Thus, in 1994, a double-blind, randomized, placebo-controlled trial (the Vasovagal Syncope International Study –VASIS)13 was started in an attempt to establish the real value of etilefrine in the treatment of vasovagal syncope. In this article, the main results of this trial are reported and discussed.
The study cohort consisted of 126 patients with recurrent vasovagal syncope; they were recruited from a total of about 2000 patients referred to the 20 participating centers from September 1994 to September 1996 for evaluation of unexplained syncope by means of head-up tilt testing. To be included in the study, patients had to fulfill the following 2 criteria: at least 3 syncopal episodes in the previous 2 years, with the last episode occurring within 6 months of enrolment and with an interval between the first and last episode >6 months; and positive response to baseline head-up tilt testing.
The diagnosis of vasovagal syncope was based, in addition to a positive tilt test result, on the exclusion of all other possible causes of syncope by means of a systematic work-up. This included, in all patients, careful history and physical examination, full neurological assessment, standard laboratory tests, supine and orthostatic blood pressure measurements, 12-lead surface ECG, bilateral bedside and upright carotid sinus massage, 24-hour Holter recording, and 2-dimensional echocardiography. Other diagnostic investigations, such as exercise stress testing, electrophysiological study, cardiac catheterization, electroencephalogram, Doppler flowmeter analysis of neck vessels, and CT scan or MRI of the brain were performed only when clinically indicated.
Exclusion criteria included moderate or severe hypertension not controlled by drugs, recent (<6 months) acute myocardial infarction or angina pectoris, class III or IV NYHA functional class, other common contraindications to the use of α-sympathetic agents, concomitant severe chronic disease (eg, diabetes mellitus, neurological disease, terminal neoplasia), use of drugs that can favor or prevent vasovagal syncope (eg, nitrates, beta-blocking agents), age <18 years, weight <50 kg, pregnancy, and refusal to participate in the study. All eligible patients were informed as to the nature of the trial and gave their written consent before enrolment.
The study was planned as a double-blind, randomized, controlled trial with parallel groups. The study protocol was approved by the Ethics Committee in each participating center. Eligible patients were assigned, according to a central computer-generated randomization list, to one of the 2 study arms: placebo or etilefrine at a dosage of 25 mg 3 times a day (1 capsule every 8 hours). Randomization was stratified by balanced blocks of 4 patients. Several blocks, sufficient to cover all patients, were assigned to each center. Etilefrine and placebo (Boehringer Ingelheim) were provided as 25 mg capsules identical in shape, color, and size.
After 3 to 6 days of treatment, a second head-up tilt test was performed in the same manner as the baseline test. Regardless of the response obtained, the patients continued with the treatment initially established and were followed up. The reason for repeating the head-up tilt test was to establish the short-term reproducibility of the test as well as its value in predicting the efficacy of long-term therapy with etilefrine (a posteriori analysis).
During follow-up, examinations were performed every 3 months. At each examination an accurate history, a physical examination and an ECG were performed; in particular, a record was kept of any syncopal or presyncopal recurrence, associated traumas, side-effects, variations in treatment from the initial dosage, and other concomitant therapy. Titration decrements in etilefrine or placebo to 25 mg twice a day (2 capsules a day) was permissible. Drug compliance was assessed by pill counting. The follow-up duration was 1 year or until syncope recurred. Predefined criteria of withdrawal from the study included major protocol violations, poor patient cooperation or noncompliance in returning for examination and/or administration of drug medication, occurrence of serious adverse events (defined as untoward medical events requiring hospitalization or producing temporary or permanent invalidation or life-threatening or fatal), pregnancy, and the patients’ refusal to continue. Patients who dropped out because of intolerable adverse experiences, lack of adequate therapeutic response resulting in intolerable signs and/or symptoms, or unacceptable risk were considered complete cases for the intention-to-treat analysis.
The primary study end-point was the first recurrence of syncope during follow-up. Secondary end-points were the recurrence of presyncope, the total number of presyncopal episodes, and the usefulness of etilefrine in preventing tilt-induced syncope (short-term study).
Head-Up Tilt Testing Protocol
Head-up tilt testing was always performed in the morning, after overnight fasting, in a quiet, slightly darkened room. The procedure was performed by means of an electronically controlled tilt table with a footboard for weight-bearing. Blood pressure was measured by means of an Ohmeda Finapress 2300 (Ohmeda) photoplethysmographic device that provides noninvasive beat-to-beat determination of finger arterial pressure by the Penaz volume-clamp method.14
Initially (first 24 patients), head-up tilt testing was performed without drug challenge, according to the Westminster protocol (passive tilting at 60° for 45 minutes).15 Beginning in March 1995, in an attempt to increase the recruitment rate, sublingual nitroglycerin provocation16 was added. If syncope did not develop during the initial unmedicated phase at 60° for 45 minutes, 300 μg of nitroglycerin was administered sublingually, and patients continued to be tilted at 60° for an additional 20 minutes. The end-points of head-up tilt testing were positive response (see below) and completion of the protocol.
Syncope was defined as sudden transient loss of consciousness, with inability to maintain postural tone, and with spontaneous recovery. Presyncope was defined as the complex of premonitory signs and symptoms of imminent syncope (eg, severe light-headedness, severe weakness, transient graying of vision, or hearing loss) with difficulty in maintaining postural tone. A positive response to head-up tilt testing was defined as reproduction of the spontaneous syncope in association with hypotension, bradycardia, or both (decrease in systolic blood pressure >50% and decrease in heart rate >30% of the maximal value observed in the upright position). During the drug-free phase of the test, the occurrence of isolated hypotension was considered sufficient to classify the response as positive. During the pharmacological phase, both hypotension and bradycardia of relatively rapid onset (occurring within 5 minutes) were required.16 The positive response to head-up tilt testing was defined as cardioinhibitory when a marked bradycardia (≤40 bpm for >10 seconds) or prolonged asystole (≥3 seconds) occurred at the time of syncope.13 The positive response was defined as vasodepressor or mixed when isolated hypotension or hypotension associated with only mild bradycardia (>40 bpm) or brief asystole (<3 seconds) was observed.13
Sample Size Calculation
The formula used to compute the sample size17 is reported in Appendix 1. The sample size calculation was based on the results of a preliminary study9 that suggested an expected 15% 1-year syncopal recurrence rate in the placebo group and an expected 5% 1-year syncopal recurrence rate in the etilefrine group. With an alpha level of 0.05 and a test power of 0.80, the resulting sample size was 137 patients for each treatment group; this was increased to 180 patients on account of the number of dropouts anticipated.
For ethical and administrative problems, formal interim analysis was planned and performed in September 1996, after the first 70 patients had concluded 1-year follow-up. This interim analysis revealed a lack of evidence of etilefrine efficacy, and it was estimated that, even with a considerably greater number of patients (360 or more), it would have been practically impossible to show any significant difference between etilefrine and placebo. Thus, the Steering Committee decided to stop recruitment.18 The final statistical analysis was performed on the 126 patients who had already been enrolled at the time of interim analysis. With this number of patients the test power is 0.50.
The primary analysis of the results of the study was by intention-to-treat but we also made an on-treatment analysis of the primary end-point in enrolled patients (with >70% compliance) while they were on study medication or within 3 days of early permanent discontinuation. The recurrence of syncope in the 2 treatment groups was tested by using the odds ratio of the 2-binomial proportion analysis. Moreover, the time to the first syncopal recurrence was analyzed by means of Kaplan-Meier curves. Means (±SD) and medians were calculated for continuous variables, and frequencies were measured for categorical variables. Differences between groups were examined for statistical significance by a 2-sample Student’s t test for continuous variables and by Fisher’s exact test for categorical variables. P<0.05 was considered significant.
The trial profile shows overall patient numbers throughout the study (Figure 1⇓). Screening logs were not maintained throughout the trial, but we used data from tilt laboratory records to calculate that, during the study recruitment period, about 2000 patients with unexplained syncope were referred for tilt testing evaluation and that about 250 of these (approximately 12%) met eligibility criteria. Of these, 126 of the eligible patients gave their informed consent to take part (≈6% of the source population). Of the 126 enrolled patients, 63 were randomly assigned to the etilefrine arm and 63 to the placebo arm. The baseline characteristics of randomized patients in the 2 study groups were broadly similar (Table 1⇓). The patients were followed up for a maximum of 1 year (mean, 262 days). During the follow-up, 13 patients (5 in the etilefrine group and 8 in the placebo group) were not assessable by on-treatment analysis owing to poor (<70%) drug compliance or >3 days’ drug discontinuation. Thus, the efficacy analysis was performed only on a total of 113 patients. Outcome events are summarized in Table 2⇓.
In the intention-to-treat analysis, the group treated with etilefrine had the same number of syncopal episodes as the placebo group (15 patients, 24%). In the on- treatment analysis, similar results were found. Median time to the first recurrence of syncope was no different between the etilefrine group and the placebo group (106 days versus 112 days). The Kaplan-Meier actuarial estimates of first recurrence of syncope by intention-to-treat analysis and by on-treatment analysis are shown in Figures 2⇓ and 3⇓, respectively. The probability of remaining free of syncope drops steadily and similarly in both groups.
No statistical difference was observed between the 2 study groups concerning the number of patients who had at least one presyncope during follow-up and the total number of presyncopal episodes. On repeated head-up tilt testing performed 3 to 6 days after the start of treatment, etilefrine did not appear superior to placebo in preventing tilt-induced syncope (negative response 52% in etilefrine group versus 45% in placebo group).
Variables Predictive of Syncopal Recurrence
In order to assess the value of some variables in predicting the risk of syncopal recurrence during follow-up, the data of all 126 patients enrolled were analyzed together (owing to the lack of significant difference between the two study groups concerning the primary end-point). Among the different variables examined, many syncopal episodes (>6) and a positive response to repeated head-up tilt testing on treatment were predictive of a higher risk of syncopal recurrence (45% in patients with >6 episodes versus 20% in patients with ≤6 episodes, P<0.04; and 32% in patients with positive response to repeated tilt test versus 15% in patients with negative response, P<0.05; Table 3⇓).
A total of 30 adverse events judged as possibly related to the treatment were registered, 17 occurring in patients treated with etilefrine and 13 occurring in the placebo group. The occurrence of adverse events was comparable between the 2 treatment groups, the only exception being psychiatric disorders, such as agitation, depression, anxiety, and insomnia, which were exclusively observed in patients treated with etilefrine (20% versus 0%). In particular, the incidence of gastrointestinal discomfort, palpitations, and new onset or worsening of hypertension, generally considered common side-effects of therapy with α-agonist agents, was not significantly higher in the etilefrine group than in the placebo group. Overall, the majority of adverse events were mild/moderate in severity. However, 7 patients (4 in the etilefrine group and 3 in the placebo group) discontinued treatment prematurely because of adverse events. No deaths or serious syncope-associated traumas occurred throughout the study.
In recent years, several investigators have evaluated the utility of α-agonist medications in the treatment of vasovagal syncope, and the reported results have usually been positive.8 9 10 11 12 However, the majority of the trials conducted with these drugs have been open studies performed in small numbers of patients and have involved different compounds, such as etilefrine, pseudoephedrine, and midodrine. To date, there are only 2 small controlled trials that have been performed with α-agonist agents with discordant results. In the Moya trial,10 no difference was observed between etilefrine and placebo, whereas in the Ward trial,12 midodrine appeared to be clearly superior to placebo. It is noteworthy that both these trials were essentially short-term studies designed to test the efficacy of α-agonist agents in preventing tilt-induced syncope. Moreover, the dosage of etilefrine in the Moya study (10 mg, 3 times a day) was probably too low to achieve a marked vasoconstrictor effect, as recognized by the same authors.10 Thus, at the present time, we have insufficiently compelling data on the real efficacy of long-term therapy with α-agonist agents for the prevention of spontaneous recurrences of vasovagal syncope. The VASIS trial was a multicenter, double-blind, randomized, placebo-controlled trial, which was planned in 1994 to answer this question. The trial was prematurely interrupted in September 1996 after the first interim analysis had been performed.
The main finding of the VASIS trial is the lack of superiority of etilefrine over placebo in preventing spontaneous syncopal recurrence. Indeed, the 63 patients treated with etilefrine at the dosage of 25 mg 3 times a day showed exactly the same recurrence rate during the first year following enrolment as did the 63 patients treated with placebo (24%). Furthermore, the time to the first syncope after the start of treatment, as well as the incidence and the number of presyncopal episodes, were not significantly different between the 2 study groups. These results confirm those of 3 previous small randomized controlled trials performed by Fitzpatrick et al,19 Brignole et al,20 and Morillo et al.21 with other drugs, and, like those results, strongly suggest caution in interpreting the outcome of open studies that do not include a placebo or control group. Indeed, the frequently benign course of vasovagal syncope, with relatively long periods of quiescence and asymptomatic status regardless of the intervention, may give the false impression of drug efficacy.
The relatively low recurrence rate of symptoms in our study population, despite the predicted high risk of relapses, is in agreement with the observations made by many authors20 21 22 and may be explained by the cyclic course of neurally mediated syncope, patient reassurance, a possible therapeutic action of repeated tilting,23 and the placebo effect of the treatment.
Like other studies,10 19 20 21 the VASIS trial documented a marked reduction in the incidence of positive responses to repeated tilt testing on treatment, whether etilefrine or placebo was administered, with no difference between the 2 study groups (52% negative conversion of a positive tilt result with etilefrine and 45% with placebo). This suggests a low short-term reproducibility of positive tilt tests and discourages the use of serial head-up tilt testing as a method for predicting the efficacy of a given therapeutic intervention.24 It is noteworthy that the persistence of a positive response to repeated tilt testing on treatment was able to identify a subgroup of patients with a higher incidence of syncopal recurrences during follow-up. Many previous syncopal episodes (>6), in accordance with the data of Sheldon et al25 and Grimm et al,26 was also predictive of a greater probability of syncopal relapse. The identification of clinical variables significantly associated with syncopal recurrences is clearly of great value in detecting those patients with vasovagal syncope who really need long-term treatment.
In conclusion, the VASIS trial has clearly demonstrated that etilefrine is not superior to placebo in preventing vasovagal syncope. Moreover, it has taught us that today only randomized placebo-controlled trials are able to assess the real efficacy of any proposed treatment for patients with vasovagal syncope. Finally, it has shown the limited value of tilt table testing in predicting a positive outcome, which must caution us in the interpretation of other studies using response to tilt table testing as therapy end-point.
Formula used to compute sample size: where p1=percentage of events expected in placebo group p2=percentage of events expected in etilefrine group q1=100−p1 q2=100−p2 Z1−α/2=standard normal distribution for error I type Z1−β=standard normal distribution for error II type
VASIS Participating Centers and Investigators
Ospedale Umberto I, Mestre-Venezia: Franco Giada, MD, Gianni Gasparini, MD; Ospedale S. Maria Nuova, Reggio Emilia: Gino Lolli, MD; Ospedale Bolognini, Seriate-Bergamo: Giuseppe Leoni, MD, Vittorio Giudici, MD; Ospedale Ferrari, Casarano-Lecce: Giacinto Pettinati, MD, Fernando De Sanctis, MD; Ospedale Civile, Mirano-Venezia: Federico Sartori, MD, Albino Zanocco, MD; Ospedale Nuovo S. Gerardo, Monza-Milano: Antonio Vincenti, MD, Sergio De Ceglia, MD; Ospedale Riuniti, Lavagna-Genova: Daniele Oddone, MD; Ospedale S. Maria delle Croci, Ravenna: Silvano Silvani, MD, Maurizio Piancastelli, MD; Ospedale Maggiore, Lodi-Milano: Mario Orlandi, MD, Egidio Marangoni, MD; Ospedale Civile, Borgomanero-Novara: Marco Zanetta, MD, Umberto Parravicini, MD; Ospedale Valduce, Como: Mauro Santarone, MD, Giovanni Foglia Manzillo, MD; Ospedale Fatebenefratelli, Roma: Massimo Fioranelli, MD; Ospedale Giuseppe Moscati, Avellino: Giuseppe De Fabrizio, MD, Francesco Rotondi, MD; Ospedale S. Orsola Fatebenefratelli, Brescia: Alessandra Marchetti, MD; Ospedale Civile, Cento-Ferrara: Maurizio Dinelli, MD; Ospedale Civile, Imperia: Giacomo Musso, MD, Antonello Ranise, MD; Ospedale Civile, Castelnuovo Garfagnana-Lucca: Eugenio Nannini, MD, Pier Romano Mariani, MD; Ospedale Silvestrini, Perugia: Pietro De Ciuceis, MD, Vincenzo Chiodini, MD; Ospedali Riuniti, Taranto: Vitantonio Russo, MD, Nicola Baldi, MD; Ospedali Riuniti, Salerno: Michele Di Mauro, MD, Andrea Campana, MD.
VASIS Data Coordinating Center: Boehringer Ingelheim Italia, Milano: Elena Cicioni, ScD; Medical Statistics Service: Mauro Montanari, PhD, Paola Ambroso, BScD.
↵1 A complete list of the Vasovagal Syncope International Study (VASIS) Investigators appears in Appendix 2.
- Received August 20, 1998.
- Revision received November 18, 1998.
- Accepted December 11, 1998.
- Copyright © 1999 by American Heart Association
Epstein SE, Stampfer M, Beiser GD. Role of the capacitance and resistance vessels in vasovagal syncope. Circulation. 1968;37:524–533.
Hargreaves AD, Muir AL. Lack of variation in venous tone potentiates vasovagal syncope. Br Heart J. 1992;67:486–490.
Morillo CA, Eckberg DL, Ellenbogen KA, Beightol LA, Hoag JB, Tahvanainen KU, Kuusela TA, Diedrich AM. Vagal and sympathetic mechanisms in patients with orthostatic vasovagal syncope. Circulation. 1997;96:2509–2513.
Raviele A, Themistoclackis S, Gasparini G. Drug treatment of vasovagal syncope. In: Blanc JJ, Benditt D, Sutton R, eds. Neurally Mediated Syncope: Pathophysiology, Investigations, and Treatment. Armonk, NY: Futura Publishing; 1996:113–117.
Moya A, Permanyer-Miralda G, Sagrista-Sauleda J, Carne X, Rius T, Mont L, Soler-Soler J. Limitations of head-up tilt test for evaluating the efficacy of therapeutic interventions in patients with vasovagal syncope: results of a controlled study of etilefrine versus placebo. J Am Coll Cardiol. 1995;25:65–69.
Ward CR, Gray JC, Gilroy JJ, Kenny RA. Midodrine: a role in the management of neurocardiogenic syncope. Heart. 1998;79:45–49.
Vasovagal Syncope International Study (VASIS) Group. Double-blind randomised study etilefrine versus placebo in patients with tilt-induced recurrent vasovagal syncope. Eur J Cardiac Pacing Electrophysiol. 1993;2:164–172.
Pocock SJ. Clinical Trials: A Practical Approach. Chichester: John Wiley & Sons; 1983(reprint 1984):123–125.
Pocock SJ. When to stop a clinical trial. BMJ. 1992;305:235–240.
Fitzpatrick AP, Ahmed R, Williams S, Sutton R. A randomised trial of medical therapy in “malignant vasovagal syndrome” or “neurally-mediated bradycardia/hypotension syndrome.” Eur J Cardiac Pacing Electrophysiol. 1991;2:99–102.
Sheldon R, Rose S, Flanagan P, Koshman ML, Killam S. Risk factors for syncope recurrence after a positive tilt-table test in patients with syncope. Circulation. 1996;93:973–981.
Grimm W, Degenhardt M, Hoffmann J, Menz V, Wirths A, Maisch B. Syncope recurrence can better be predicted by history than by head-up tilt testing in untreated patients with suspected neurally mediated syncope. Eur Heart J. 1997;18:1465–1469.