Syncope, defined as a transient loss of consciousness due to transient global cerebral hypoperfusion, characterized by rapid onset, short duration, and spontaneous complete recovery, is a frequent symptom and can be attributed to different causes.1
The most common aetiology of syncope is a neurally mediated reflex that consists of a transient failure of the cardiovascular mechanisms that normally control haemodynamic stability. This reflex is usually triggered by different stimuli (afferent pathways), leading to sudden withdrawal of sympathetic activity and to an increase in parasympathetic nerve tone (efferent pathways), causing vasodilatation, with consequent hypotension, and bradycardia.
The clinical presentation of neurally mediated syncope is not usually uniform; however, it may manifest with different patterns. In some patients, it is initiated by clear adrenergic stimuli such as fear, pain, emotional distress, or medical instrumentation, whereas in others it develops after prolonged orthostatism, hypovolaemia, or carotid sinus stimuli. There are also many patients in whom reflex syncope develops without any apparent recognizable trigger. On the other hand, some patients complain of typical vegetative prodrome, whereas others develop sudden syncope without any warning symptom. In addition, although both mechanisms, hypotension and bradycardia, are, by definition, always present, the predominance of one or the other leads to the characteristic patterns of vasodepressor, mixed, or cardioinhibitory reflex syncope.2 Although all these situations share a common underlying mechanism, i.e. hypotension and bradycardia triggered by some afferent stimuli, these differences have a great impact on the diagnostic process and in therapeutic decisions. Whereas a syncopal episode in a young patient, with clear triggers and prodrome, is easy to diagnose and manage, sudden episodes, without recognizable triggers, particularly in the older population, represent a diagnostic challenge and are usually difficult to treat.
In 23–50% of patients, the aetiological diagnosis of syncope can be achieved with initial evaluation consisting of clinical history, physical examination, ECG, blood pressure measurement in the supine position and orthostatism, and, in those >40 years, carotid sinus massage.1 In most of the remaining patients, a normal physical examination and ECG, and absence of structural heart disease, family history of sudden death, or data suggestive of cardiac syncope, such as palpitations or syncope during exercise, render the diagnosis of neurally mediated syncope highly suggestive and no other tests are usually required. There are, however, some patients in whom, owing to very frequent recurrent episodes, injury associated with syncope, or occupational implications, every effort should be made to achieve the most accurate diagnosis, not only to rule out other causes but also to select the most appropriate treatment.
Tilt testing was first described as a diagnostic tool in patients with syncope of unknown origin in 1986.3 In that first description, a passive test with a tilt angle of 40° for 60 min was used. Since then, many authors have published several different protocols to improve the diagnostic yield and find even shorter and simpler protocols. In 1989, a protocol using incremental doses of intravenous isoproterenol was described.4 The rationale for using isoproterenol was based on the observation that some reflex episodes were preceded by an increase in adrenergic tone manifested by tachycardia and an increase in blood pressure. In 1995, Morillo et al. published the ‘low dose’ intravenous isoproterenol protocol which, while maintaining the same positivity rates, improved specificity.5 In 1994, Raviele et al. described a protocol using intravenous nitroglycerin aimed at increasing the venous pooling provoked by passive tilting.6 In 1995, they proposed the use of sublingual nitroglycerin.7 Since this protocol did not require the insertion of a venous line, it maintained the rate of positive responses, thereby improving specificity.7 Since then, many authors have been seeking the best protocol for different specific populations, mainly to modify the duration of the test or even obviate the initial unmedicated passive phase.8
In addition to the problem of the co-existence of many different protocols, some other limitations arise when interpreting the results of tilt testing in patients with syncope of unknown origin. The first is that although all these protocols have relatively high specificity, the true sensitivity remains unknown because of the lack of a reliable ‘gold standard’ for patients with established neurally mediated syncope.5–8 Moreover, the reproducibility of the test is relatively low,9 thereby limiting its use for evaluation of the response to treatment and predicting follow-up. On the other hand, recent publications compared the response to different tilt testing protocols with the syncopal recurrence rate and heart rate behaviour during spontaneous syncope, assessed by an implantable loop recorder.10,11 Whereas a positive cardioinhibitory response to the tilt test predicts an asystolic spontaneous syncopal episode with high probability (75–80%),10 a negative test or a positive vasodepressor or mixed response do not rule out the presence of recurrent syncopal episodes with severe asystole.10,11 These data confirm that the sensitivity of tilt testing is probably lower than expected and that its use for assessing the exact mechanism, and eventually for selecting therapy in patients with severe or recurrent syncope episodes, has limitations.
For many years, Theodorakis et al. have proposed the use of clomipramine, a serotonin re-uptake inhibitor, during tilt testing to increase the diagnostic yield of the test.12,13 The rationale for its use was based on the observed role of serotonin in triggering neurally mediated syncopal episodes.14 Flevari et al.15 published the results of a crossover randomized study that compared tilt testing potentiated with intravenous clomipramine with tilt testing potentiated with sublingual nitroglycerin in 100 patients with suspected neurally mediated syncope and in 50 asymptomatic control subjects. The main finding of that study was that clomipramine significantly increased the rate of positive response to tilt testing compared with nitroglycerin in patients with previous syncope (73% with clomipramine and 52% with nitroglycerin); however, no differences were observed in the control group (10% vs. 12%). These data suggest that clomipramine increases the sensitivity of the test without compromising its specificity.
How should we interpret these results and how should they be placed in a clinical context?
These results add to the understanding of the mechanisms of reflex syncope. They confirm that neurally mediated syncope can not only be provoked by increased sympathetic nerve tone due to stressful stimuli or by hypovolaemia or orthostatism, but can also be initiated by some central nervous system triggers. This complexity may explain why different drugs, such as isoproterenol, nitroglycerin, or clomipramine, acting at very different levels, can trigger a neurally mediated reflex during tilt testing (Figure 1).
Does this mean that clomipramine tilt testing should replace currently used protocols? Before recommending implementing a new protocol in the diagnostic work-up of patients with syncope, several aspects should be adressed. One is to what extent this protocol predicts outcome and heart rate pattern during spontaneous syncope. Moreover, instead of suggesting that the clomipramine test should replace currently recommended protocols, the question relies on the need to select a specific protocol according to the clinical characteristics of syncopal presentation, such as, for example, the nitroglycerin test for patients with a predominant orthostatic trigger, the isoproterenol test for those with syncope mainly triggered by adrenergic stimuli, and perhaps the clomipramine test for those with suspected central nervous system trigger or without apparent trigger. A further important aspect, as stressed by the authors, is that the use of clomipramine in tilt testing to date has been mainly reported by the same group,12–15 and these results should be confirmed by other groups or by multicentre trials.
Conflict of interest: none delared.