Is There an Effective Treatment for Neurally Mediated Syncope?

During the past 20 years, a great deal has been learned about the evaluation, management, and prognosis of syncope. Studies from the 1980s showed that for approximately half of patients with syncope, the cause could not be established using diagnostic tests widely available at the time.¹ Studies of risk stratification showed that the most important predictor of a poor outcome was underlying structural heart disease, including abnormalities detected on electrocardiogram.^{2 - 3} These findings prompted investigators to devise new diagnostic tests and to propose alternative strategies for managing syncope.

In the subset of patients with syncope and structural heart disease, the focus shifted to extensive evaluation of cardiovascular disease and to the use of electrophysiological studies to test for arrhythmias. In the subset of patients who have syncope, but in whom structural heart disease or other cardiac etiology could not be diagnosed and did not seem clinically plausible, the focus shifted to the use of new diagnostic studies, such as tilt testing, prolonged event monitoring using external and insertable loop recorders, and to the use of psychiatric and psychological evaluation. In this second subset, findings have shown that about 65% of patients have positive results on tilt testing,⁴ whereas smaller percentages have brief arrhythmias (mostly bradyarrhythmias) that are detectable by monitoring with loop recorders^{1 ,5 - 6} or have a psychiatric illness,⁷ such as panic disorder, anxiety disorder, depression, or somatization disorder.

A large part of the effort has focused on understanding neurally mediated syncope induced by tilt testing and its management. Three types of responses to this procedure are considered abnormal: (1) neurally mediated response, which is characterized by the sudden onset of hypotension, bradycardia, or both when the patient is kept in an upright position; (2) postural orthostatic tachycardia syndrome, which is characterized by tachycardia that begins when the patient is placed in the upright position on the tilt table and persists throughout the procedure; and (3) dysautonomia, which is characterized by a gradual decrease in blood pressure with little or no change in heart rate during the procedure.⁸ Almost all of the literature on syncope induced by tilt testing has centered on neurally mediated syncope, which is the most common response during tilt testing.

Because there is no criterion standard for the diagnosis of neurally mediated syncope, it is appropriate to question how reliable tilt testing is for detecting this condition. Tilt testing has been found to have a sensitivity of 67% to 83% when patients with clinically diagnosed vasovagal syncope are tested.⁹ The specificity of this test varies with technique. For example, the specificity of the procedure is about 90% when it is used with a low-dose isoproterenol protocol or nitroglycerin, but it is lower when used with a high-dose isoproterenol protocol.^{4 ,10 - 11} Thus, medical centers that use highly aggressive protocols may be at increased risk of making false-positive diagnoses. However, because there is no standardized approach to tilt testing and because many variations of the procedure are used clinically and described in the literature, it is difficult to compare results from one medical center with those of another.

A number of issues concerning the treatment of neurally mediated syncope have arisen. First, neurally mediated syncope has varied clinical manifestations. Situational syncope (occurring after micturition, defecation, swallow, or cough), emotional faint (occurring in response to pain or stress), or carotid sinus syncope as well as syncope triggered by use of certain drugs (eg, nitroglycerin) or by volume depletion are all considered different manifestations of neurally mediated syncope. Treatment of these entities is markedly different than the management of unexplained syncope with a positive tilt test result.

Second, neurally mediated syncope may affect patients' lives differently. On one extreme, syncope is a rare or isolated event and may not result in injury. In a small subset of patients, however, syncope is a frequent event (such as monthly or more) and places the patient at risk for injuries, motor vehicle crashes, and disability that leads to loss of work, depression, and despair. The focus of almost all treatment studies has been on syncope that occurs relatively frequently and is likely to have an effect on patients' lives. This is appropriate because there are no objective ways for determining the benefit of treatment in patients who have one episode or rare episodes. Additionally, it would be difficult to recommend a daily medication or permanent pacemaker to prevent syncopal episodes that have an extremely low likelihood of occurring. Expert consensus statements have in fact recommended against treating patients with rare episodes.⁸

Third, the effectiveness of drug therapy to prevent recurrent neurally mediated syncope is open to question. Measures such as volume expansion with increased salt and fluid intake, moderate exercise, and tilt training are relatively safe, but their effectiveness has not been demonstrated by controlled trials. One small, short-term (1-month) randomized trial of β-blockers and a large number of uncontrolled studies of β-blockers claim effectiveness of these drugs, but several controlled trials did not show β-blockers to be effective.^{12 - 15} In a randomized trial of etilephrine, this vasoconstrictive agent was found to be ineffective.¹⁶ Other drugs have not been studied extensively in controlled trials. In a crossover trial of midodrine and placebo in 16 patients, treatment for 1 month with midodrine was found to alleviate symptoms and improve the quality of life.¹⁷ In a study comparing treatment with midodrine to increased salt and fluid intake in 61 patients, midodrine was found to alleviate symptoms at 6 months.¹⁸ A randomized trial of paroxetine in 41 patients showed a reduction of recurrent episodes at 2 years.¹⁹ Larger controlled studies with longer duration of follow-up are needed to define the role of these and other drugs in the treatment of neurally mediated syncope.

The rationale for cardiac pacemaker implantation is that some degree of bradycardia is frequently noted during vasovagal episodes that are induced by tilt testing or that occur spontaneously. On tilt testing, approximately 65% of patients with syncope have cardioinhibitory responses.⁴ Atropine fails to prevent syncope despite abolishing bradycardia, suggesting that vasodilatation plays a more prominent role in causing syncope than does bradycardia. Uncontrolled studies of pacemakers have suggested that these devices may ameliorate symptoms or delay the development of syncope on tilt testing.²⁰ To improve the efficacy of pacing, investigators postulated that the likelihood of preventing syncope would be greater if pacemakers could sense the onset of vasovagal syncope and provide sufficient and early heart rate support to overcome transient hypotension that accompanies neurally mediated syncope. These concepts led to using rate hysteresis and rate-drop algorithms with pacemakers in the treatment of patients with neurally mediated syncope.

The effectiveness of permanent pacemakers has been assessed in 3 previously reported randomized trials. The North American Vasovagal Pacemaker Study (VPS) randomized 54 patients to receive a pacemaker or no pacemaker.²¹ Patients had experienced 6 or more episodes of syncope during their lifetime and had positive tilt test results with some degree of bradycardia (heart rate <60/min when isoproterenol was not used and <70/min or <80/min when isoproterenol was used at 2 different maximal dosages). In 19% to 26% of patients in this study, the lowest heart rate recorded was less than 40/min. The mean age was 40 years in those who received a pacemaker and 46 years in those who did not. Syncope recurred in only 22% of patients with pacemakers (representing an 85.4% reduction in the relative risk for recurrence), but recurred in 70% of patients without pacemakers.

In the Vasovagal Syncope International Study (VASIS), patients were randomized to receive a pacemaker with rate hysteresis or no pacemaker.²² The study included 42 patients who had experienced 3 or more episodes of syncope during the past 2 years and had a cardioinhibitory response on tilt testing (almost all without drug provocation), which was defined as a heart rate of less than 40/min for more than 10 seconds or asystole for more than 3 seconds. The mean age was 64 years in those who received a pacemaker and 56 years in those who did not. During a mean follow-up period of 3.7 years, syncope recurred in 5% of patients with pacemakers and 61% of patients without pacemakers.

In the Syncope Diagnosis and Treatment Study (SYDIT), 93 patients were randomized to receive a pacemaker with rate-drop response or atenolol.²³ Patients had 3 or more episodes of syncope during the past 2 years and a positive tilt test (using nitroglycerin provocation) result with relative bradycardia (minimum heart rate <60/min). The mean age was 61 years in those who received a pacemaker and 55 years in those who received atenolol. Asystole was noted on tilt testing in 56 patients. Syncope recurred in 4.3% of patients who received pacemakers and 25.5% of patients who received atenolol (at a median of 390 days).

The Second Vasovagal Pacemaker Study²⁴ (VPS II) reported in this issue of THE JOURNAL has gone one step further in addressing the issue of the placebo effect associated with pacemaker intervention. In this study, patients from several different medical centers were randomized to receive a pacemaker with rate-drop response (DDD) or a pacemaker with sensing but without pacing (ODO). The study included 100 patients with a mean age of 51 years in the DDD group and 48 years in the ODO group. Patients had 6 or more episodes of syncope during their lifetime or had at least 3 episodes during the past 2 years. There was no standardized protocol for tilt testing, and the centers participating in the study used isoproterenol or nitroglycerin during the testing. In only 15% to 23% of the patients, the lowest heart rate recorded was less than 40/min. After a 6-month follow-up, the study showed a 30% risk reduction in the recurrence of syncope, which was not statistically significant. The study results raise the question of whether pacemakers have any role in the treatment of neurally mediated syncope.

In the 3 earlier studies, there was some concern about whether the reduction in the risk for syncope recurrence may have been due to a placebo effect of the pacemaker. Because implantation is a major intervention, patients with vasovagal syncope experience emotional and psychological factors, which may modulate reflex syncope and the autonomic response to tilt testing. The VPS II trial addressed the possibility of the placebo effect by inserting pacemakers in both groups and blinding patients, study personnel, and physicians to the randomization. These measures addressed the major criticisms of prior studies.

There are several reasons that it may be prudent to avoid pacemakers in the treatment of neurally mediated syncope. First, because vasodilatation is such a central part of this condition, there seems to be little physiological foundation for the belief that pacemakers are effective in preventing the recurrence of syncope. Second, since recurrent episodes of syncope are often clustered, a permanent form of treatment, such as a pacemaker, may be less attractive than a less invasive form of treatment that is directed toward managing syncopal episodes over a brief period. Many patients become relatively or entirely free of symptoms after having a great deal of episodes over a short period. Third, many patients with neurally mediated syncope are young and otherwise healthy. Inserting a permanent pacemaker means that they will need continued surveillance and may experience discomfort and inconvenience for decades.

Is there any role for pacemakers in patients with neurally mediated syncope? A comparison of the characteristics of the patient populations of the 4 pacemaker trials raises the question of whether the differences in the study results could be due to differences in the study populations or in tilt testing methods. In VPS and VPS II, only a small percentage of patients (<30%) had relatively severe bradycardia. In VASIS and SYDIT, patients tended to have more severe bradycardia. Moreover, 85% of patients in VASIS and 60% of patients in SYDIT had asystole. In the case of neurally mediated syncope, it is possible that relative bradycardia on tilt testing is not an important predictor of the response to pacemakers but that asystole is. While tilt testing in VASIS and SYDIT was reasonably standardized, it was not standardized in VPS or VPS II. Whether the lack of standardization had an impact on how patients were classified as having neurally mediated syncope is impossible to determine. Finally, patients in VASIS and SYDIT were older, which raises the question of whether neurally mediated syncope in older individuals has a different pathophysiological mechanism requiring different treatments.

During the past 2 decades, some progress has been made in the diagnosis and management of neurally mediated syncope. A potential cause of syncope can now be determined in many patients whose condition would earlier have been classified as unexplained syncope. However, in neurally mediated syncope, which is the most common form of syncope, the benefits of making a diagnosis are not yet realized. The results of most drug treatment trials have been disappointing, and no clearly effective drug has emerged. The value of implanting pacemakers has been repeatedly questioned, and now there is evidence of their lack of effectiveness. Major breakthroughs are not likely to come from more trials of existing drugs or from a better definition of which types of patients may benefit from pacemakers. The breakthroughs are likely to come from a better understanding of the pathophysiological and molecular mechanisms of neurally mediated syncope. Insights into these mechanisms will offer the best hope of developing new and effective methods to treat patients with this potentially disabling condition.