Head-up Tilt Test
A Nonspecific Method of Evaluating Patients With Bifascicular Block
Background Patients with bifascicular block have an increased risk of syncopal attacks, but the underlying mechanism often remains unclear despite an extensive diagnostic workup. The head-up tilt test has been established as an important diagnostic tool in the unmasking of vasovagal syncope in patients with unexplained syncope. Its role in the evaluation of patients with bifascicular block has not been studied.
Methods and Results A head-up tilt test, using a 60° angle of tilt for 45 minutes without pharmacological provocation, was performed in 25 patients with bifascicular block and syncope that remained unexplained after an extensive invasive and noninvasive electrophysiological investigation. As a control group, 25 subjects with bifascicular block without syncope, matched for age, sex, left ventricular function, and underlying heart disease, were included. A positive head-up tilt test was found in 7 (28%) of the syncope patients and in 8 (32%) of the control subjects (P=NS). Six patients, of whom 3 had a positive tilt test, had recurrent syncopal attacks during 32 months of follow-up. None of the control subjects had syncope during follow-up.
Conclusions This study gives rise to serious concern regarding the specificity of the head-up tilt test in patients with bifascicular block. A head-up tilt test should therefore be interpreted with caution, and its role as a diagnostic tool in this patient category remains to be established.
Bifascicular block, defined as left bundle-branch block or right bundle-branch block with left anterior or posterior fascicular block, has a prevalence of 1% to 1.5% in an adult population.1 Up to 25% of patients with bifascicular block have a history of syncope,1 and the management of these patients represents a common clinical problem. It has been shown that the mechanism behind the syncopal attacks remains unclear in up to 70% of patients, despite an extensive diagnostic workup.2 3 4 It could be hypothesized that a large proportion of these patients have vasovagal syncope that cannot be clinically diagnosed.
Since its introduction into clinical practice in 1986, the head-up tilt test has become a widely used method in the evaluation of patients with unexplained syncope.5 The incidence of a positive tilt test in patients with unexplained syncope has ranged between 26% and 90%, depending on the protocol used.6 The specificity, although less well studied, is ≈90%.
The objective of the present study was to prospectively examine the prevalence of tilt-induced, neurally mediated syncope in patients with bifascicular block and syncope of unknown origin. We also aimed to study the specificity of the head-up tilt test in a group of sex- and age-matched controls with bifascicular block but without syncope.
Between March 1992 and April 1994, 50 individuals with bifascicular block were prospectively studied. Twenty-five patients had a history of syncope that remained unexplained after an extensive invasive and noninvasive investigation, as outlined below. As a control group, 25 age- and sex-matched subjects with bifascicular block without a history of syncope or dizzy spells were included. They were identified with the use of a computerized storage unit for 12-lead ECGs. All subjects and patients gave informed consent in accordance with the guidelines of the Ethics Committee of the Karolinska Hospital, which had approved the study.
Investigations Before Tilt Test
Before the tilt test, all patients and control subjects underwent careful history-taking, a physical and neurological examination, a symptom-limited exercise test, and an echocardiographic Doppler examination including assessment of the left ventricular ejection fraction.
All syncopal patients had a negative 24-hour ambulatory ECG, defined as absence of sinus pauses >3 seconds, episodes of high-degree AV block, or ventricular tachycardia >5 beats. Carotid sinus massage was performed for 5 seconds on both sides, starting on the left side, and was considered abnormal if a pause >5 seconds in duration appeared.
An electrophysiological study was performed in all patients in the syncope group. No patient had a positive invasive electrophysiological study, defined as: (1) an HV interval >70 ms or His-Purkinje block during incremental atrial pacing in the basal state; (2) a positive disopyramide test, defined as an HV interval prolongation >50% or His-Purkinje block during incremental atrial pacing after intravenous disopyramide7 ; or (3) inducible monomorphic ventricular tachycardia >30 seconds in duration using 1 to 3 extrastimuli at 100 and 150 bpm at two right ventricular sites.
Other diagnostic tests, such as an electroencephalogram, head CT scans, and glucose tolerance test, were not performed unless the history or physical examination indicated specific diagnoses.
After 4 hours of fasting, all individuals were tested on a tilt table with a foot plate support. Restraining belts were placed at chest and thigh levels. No individuals underwent tilt testing while taking β-adrenoceptor blocking agents, disopyramide, nitrates, or medication with anticholinergic properties. All tilt tests were performed on a separate day from the invasive electrophysiological study. The tests were preceded by 10 minutes of observation in the supine position. A 60° angle of tilt was used, and the planned duration was 45 minutes in accordance with the Westminster protocol.8 Pharmacological provocation with isoproterenol or nitroglycerin was not used in any of the cases, and no patient had an intravenous cannulation before or during the test. Blood pressure was monitored continuously (Finapres), as was the surface 12-lead ECG. A positive test was defined as syncope or severe presyncope (state of light-headedness associated with at least one of the following symptoms: partial loss of postural tone, decreased vision, slow response to verbal stimuli, or nausea) accompanied by marked hypotension (systolic blood pressure <80 mm Hg) or bradycardia <40 bpm or both. All positive tests were classified as a cardioinhibitory, vasodepressive, or mixed reaction according to a proposal by Sutton et al.9
Treatment and Follow-up
According to the study protocol, pacemaker treatment was recommended for all patients with a history of syncope. This was accepted by 19 and declined by 6 patients. All of the implanted devices were VVI pacemakers with a bradycardia-detecting function. The lower rate was programmed to 30 bpm, which allows disclosure of the number of episodes with a spontaneous rate <30 bpm for 6 seconds via internal telemetry.10 None of the syncope patients or subjects in the control group received any therapy because of a positive tilt test. All subjects and patients were seen at the outpatient clinic after 6 weeks and thereafter, on average, every 6 months.
Left bundle-branch block and right bundle-branch block: Standard definitions were used.11
Left anterior fascicular block: Mean frontal QRS axis <−30°.
Left posterior fascicular block: Mean frontal QRS axis >90° in the absence of right ventricular hypertrophy.
Syncope: Sudden, complete loss of consciousness and muscular tone with full spontaneous recovery.
Comparisons of continuous, normally distributed variables were made by use of the Student's t test. The Mann-Whitney test was used for data that were not normally distributed. Fisher's exact test was used for comparisons of proportions. The “exact” 95% CIs of proportions were obtained from the Geigy Scientific Tables.12 Specificity was defined in the control group as the percentage of subjects with a negative test. All analyses were performed with the use of the statistical package of JMP (version 3.0, SAS Institute). A value of P<.05 was considered statistically significant.
The two groups were well matched, and there were no differences in age, sex, underlying heart disease, left ventricular function, or type of bifascicular block (Table⇓).
Result of Head-up Tilt Test
There was no difference between the syncope and the control groups regarding the proportion of patients with a positive test. In the syncope group, seven patients (28%; 95% CI, 12% to 49%) had a positive tilt test, compared with eight in the control group (32%; 95% CI, 15% to 54%) (P=NS). All seven patients with a positive test in the syncope group equated the symptoms experienced during the tilt test with those experienced during their clinical syncopal attacks. The specificity of the tilt test, measured in the control group, was 68% (95% CI, 46% to 85%). The mean time (±SD) for a positive test was 33±11 minutes in the syncope group and 22±8 minutes in the control group (P=.05). The proportion of patients with a vasodepressive (n=6), cardioinhibitory (n=6), or mixed response (n=3) was similar in the two groups. The mean (±SD) blood pressure and heart rate at the end of the tilt test were 52±19 mm Hg and 80±14 bpm in patients with a vasodepressive response. The corresponding figures were 64±30 mm Hg and 32±7 bpm and 42±14 mm Hg and 51±18 bpm in patients with a cardioinhibitory and a mixed response, respectively.
There were no differences between individuals with a positive and negative tilt test regarding age, sex, number of syncopal attacks in patients with a history of syncope, supine systolic blood pressure, ejection fraction, underlying heart disease, or type of bifascicular block.
Treatment and Follow-up
During a mean follow-up period of 32 months (range, 20 to 45 months), six patients in the syncope group had at least one recurrent syncopal attack. Five of these had a pacemaker with a bradycardia-detecting function, but bradycardia (defined as intrinsic ventricular rate <30 bpm for >6 seconds) was not detected in any of the patients. Three of the six patients with syncope during follow-up had a positive tilt test, two had a vasodepressive response, and one had a mixed response.
The control group was followed up for a mean of 30 months (range, 20 to 39 months), and no subject had syncope or presyncope during follow-up.
One of the four patients with a positive tilt test of the cardioinhibitory type had a bradycardia (>6 seconds of intrinsic heart rate <30 bpm) detected by the pacemaker during follow-up. However, he had no syncope or presyncope during follow-up. No other patient with a positive tilt test had a bradycardia documented during follow-up.
The main objective of the present study was to evaluate the prevalence of tilt-induced neurally mediated or vasovagal syncope in patients with bifascicular block and unexplained syncope. We found that the incidence of a positive head-up tilt test in patients with bifascicular block and unexplained syncope was similar to that which has been reported in other groups of syncope patients.13 However, the most important finding was that there was no difference in the incidence of positive tilt tests when patients with syncope were compared with matched control subjects. This gives rise to serious concern about the interpretation of the head-up tilt test in patients with bifascicular block.
Specificity of the Tilt Test
Several factors have been suggested to decrease the specificity of the head-up tilt test: (1) use of pharmacological provocation such as isoproterenol14 ; (2) saddle support8 ; (3) intravenous cannulation15 ; and (4) angle of tilt >60°.6 In the present study, none of these factors could have influenced the result. Moreover, patient characteristics, such as age, may influence the outcome of the tilt test.6 13 16 17 To avoid these confounding factors, we formed a control group of asymptomatic subjects with bifascicular block, all of whom remained asymptomatic during a 30-month follow-up and had demographic and clinical characteristics that were very similar to the patients in the syncope group. We found that the incidence of a positive tilt test in the control group was equally high in the syncope group, with a corresponding specificity of only 68%.
This finding is in contrast to previously reported studies in which the specificity of the head-up tilt test, performed without pharmacological provocation, has been >90%. In four of these studies,8 13 18 19 a total of 95 age- and sex-matched control subjects with a mean age >50 years were included. Despite the fact that the planned duration of the tilt was longer than in the present study in three of these studies, the specificity was 96%. A 95% CI of this figure is 88% to 98%. The corresponding figure in the present study is 46% to 85%, ie, a statistically significant lower specificity.
Why Is the Specificity Lower in Patients With Bifascicular Block?
The precise pathophysiological mechanisms of neurally mediated syncope are not completely understood. It is believed that the mechanism behind this type of syncope represents a clinical equivalent of the Bezold-Jarisch reflex, ie, a stimulation of ventricular receptors that inhibits sympathetic activity and augments cardiac vagal drive.20 This stimulation may be induced mechanically by a reduction in venous return that forces the myocardium to contract vigorously with underfilled cardiac chambers. Left bundle-branch block has been shown to impair the systolic and diastolic left ventricular function both when assessed in patients with intrinsic bundle-branch block21 as well as in right ventricular pacing–induced left bundle-branch block.22 Theoretically, this would cause less activation of the mechanoreceptors in the ventricles than a normal ventricular activation pattern would and would thereby make the patients less prone to neurally mediated syncope. This hypothesis was tested by Petersen et al,23 who performed passive tilt testing in 11 patients with cardioinhibitory vasovagal syncope, with and without atrial synchronous ventricular–inhibited (VDD) pacing in a randomized order. Unexpectedly, there was no difference in efficacy between the paced and nonpaced patients, and instead, there was a tendency toward a shorter time from onset of symptoms to syncope and a lower blood pressure at the time of symptoms when the patients were VDD paced.
Sensitivity of the Tilt Test
The proportion of patients with a positive head-up tilt test in the present study was 28%. This is in the lower range of what has been found in previous studies without pharmacological provocation.6 24 Five of the six patients with syncope during follow-up had a bradycardia-detecting pacemaker. In none of the patients was a bradycardia <30 bpm found, which indicates that other mechanisms, such as vasovagal reactions, caused the syncope. Three patients had a negative tilt test, indicating that the sensitivity of the tilt test protocol might have been too low. The use of provocative agents, such as isoproterenol, edrophonium, and nitroglycerin, would probably have increased the number of patients with a positive tilt test,13 25 26 although the specificity of these protocols has been questioned.14
The recurrence rate of syncope during follow-up was low in the syncope group, 24% during 32 months of follow-up. This is, however, similar to previous reports27 28 29 in which the annual incidence of syncope after a nondiagnostic electrophysiological study, as well as after a positive head-up tilt test, has varied between 8% and 56%.
It can be argued that the sample size of the present study population was too small. On the basis of results from previous studies, we estimated a positive rate of 40% in the syncope group and 10% in the control group. From this, we calculated that a sample size of 25 syncope patients and 25 control subjects would be sufficient to achieve a power >90% at an α-level of 0.05.
One could speculate that our control subjects had a predisposition to vasovagal syncope but did not develop symptoms at the time of the tilt test. We tried to avoid this risk by following up all patients for 30 months. This time period may have been insufficient, because previous studies29 as well as the present one have shown that the annual incidence of vasovagal syncope has been low even in untreated patients.
This study gives rise to serious concern regarding the specificity of the head-up tilt test in patients with bifascicular block and has implications with regard to the diagnostic evaluation of patients with bifascicular block and syncopal attacks. Because the risk of obtaining a false-positive test is substantial, this could lead to incorrect treatment. The results of a head-up tilt test in patients with bifascicular block should therefore be interpreted with great caution.
This study was supported by grants from the Swedish Heart and Lung Foundation and from the Karolinska Institute.
- Received June 19, 1996.
- Revision received September 16, 1996.
- Accepted October 5, 1996.
- Copyright © 1997 by American Heart Association
McAnulty JH, Rahimtoola SH. Bundle branch block. Prog Cardiovasc Dis. 1983;26:333-354.
Englund A, Bergfeldt L, Rosenqvist M. The disopyramide stress test: a sensitive and specific tool for predicting impending high-degree atrioventricular block in patients with bifascicular block. Br Heart J. 1995;74:650-655.
Sutton R, Petersen M, Brignole M, Raviele A, Menozzi C, Giani P. Proposed classification for tilt induced vasovagal syncope. Eur J Card Pacing Electrophysiol. 1992;2:180-188.
Criteria Committee of the New York Heart Association. Nomenclature and Criteria for Diagnosis of Diseases of the Heart and Great Vessels. Boston, Mass: Little Brown & Co; 1973:238-239.
Introduction to statistics, statistical tables, mathematical formulae, 2. In: Geigy Scientific Tables. Basel, Switzerland: CIBA-Geigy Ltd; 1982:85-98.
Kapoor WN, Brant N. Evaluation of syncope by upright tilt testing with isoproterenol: a nonspecific test. Ann Intern Med. 1992;116:358-363.
McIntosh SJ, Lawson J, Kenny RA. Intravenous cannulation alters the specificity of head-up tilt testing for vasovagal syncope in elderly patients. Age Ageing. 1994;23:317-319.
Grines CL, Bashore TM, Boudoulas H, Olson S, Shafer P, Wooley CF. Functional abnormalities in isolated left bundle branch block: the effect of interventricular asynchrony. Circulation. 1989;79:845-853.