Recent studies have shown conflicting data regarding left ventricular (LV) function in patients with neurocardiogenic syncope, with some investigators reporting a marked decrease in LV end-systolic wall stress and stress-corrected fractional shortening. We sought to determine the characteristics of resting LV deformation in patients with neurocardiogenic syncope by selective motion tracking of subendocardial and subepicardial regions using speckle tracking echocardiography. We assessed resting LV function in 82 patients undergoing head-up tilt-table (HUTT) testing. Patients were divided into 3 groups based on a positive HUTT test with associated co-morbid conditions (n = 30), no associated co-morbid conditions (n = 30), or negative HUTT results (n = 22). LV longitudinal, circumferential, and radial strains were obtained by speckle tracking echocardiography of subendocardial and subepicardial regions in each group and compared with resting LV deformation in 20 healthy control subjects. Compared with endocardial longitudinal strain in control subjects, that in patients with positive HUTT results was attenuated, irrespective of co-morbid conditions (p <0.05). Circumferential and radial strains did not differ among groups. On multivariate logistic regression analysis, endocardial longitudinal strain was an independent predictor (odds ratio, 1.16; p = 0.01) of positive HUTT test results. In conclusion, resting LV longitudinal strain is attenuated in patients with positive HUTT test results. Overall, these results may suggest that an increase in resting LV contractility is not a prerequisite for development of neurocardiogenic syncope.
Neurocardiogenic syncope is a common condition, accounting for 22% to 60% of all cases of unexplained syncope. The reported prevalence of syncope is between 15% and 23%, and it accounts for 3% of all emergency department visits and up to 6% of all hospital admissions. Despite the prevalence and associated morbidity of syncope, its mechanisms are not well understood, with a resulting lack of development of effective therapies. The most widely accepted hypothesis, the “ventricular theory,” proposes that the initiating event is an increase in left ventricular (LV) contractility triggered by factors such as hypovolemia due to venous pooling in the lower extremities, increased sympathetic discharge from other causes, or an underfilled ventricle. This increase in contractility leads to the activation of LV mechanoreceptors, which triggers a neural reflex (the Bezold-Jarisch reflex), resulting in a usually self-terminating episode of systemic hypotension marked by bradycardia and peripheral vasodilation. Multiple investigators have tried to assess the LV myocardial response in persons with syncope, with varying and contradictory results. We hypothesized that there may be some alteration of resting myocardial contractility in patients with neurocardiogenic syncope as diagnosed by positive results on the head-up tilt-table (HUTT) test. We assessed the LV mechanics by strain imaging by speckle tracking echocardiography, which has emerged as a sensitive tool to objectively measure myocardial contractility.
Methods
This retrospective study involved the review of clinical history, laboratory data, tilt test results, and echocardiograms. Patients referred to the cardiac electrophysiology laboratory at Mayo Clinic, Scottsdale, Arizona, for HUTT testing as part of their evaluation of unexplained syncope or presyncope between December 2007 and January 2010 were eligible for inclusion. Of 108 patients identified, 26 patients were excluded from the study for various reasons: 11 patients did not have a baseline echocardiogram, 6 patients had poor echocardiographic images, 8 patients had LV wall motion abnormalities (all 8 had myocardial infarction, and 4 each underwent subsequent percutaneous coronary intervention or coronary artery bypass surgery), and 1 patient had hypertrophic cardiomyopathy. Thus, the study group consisted of 82 patients who had undergone HUTT testing and had an assessable baseline transthoracic echocardiogram. The control group consisted of 20 healthy control subjects without known medical problems and with a normal transthoracic 2-dimensional Doppler echocardiogram.
Before the HUTT test, all patients completed informed consent documents and fasted for 4 hours. Positive response (with or without the use of isoproterenol) was defined as either sudden loss of consciousness or development of presyncope, with an abrupt decrease in systolic blood pressure (to 80 mm Hg) and reproduction of clinical symptoms. For the 15-minute control phase of the test (stage 1), each patient was placed in the supine position on the tilt table, with an intravenous catheter inserted into a peripheral arm vein and under continuous electrocardiographic monitoring. Next (stage 2), the patient was tilted upright at a 60° angle and maintained in this position for 30 minutes. If there was a positive response, the patient was returned to the supine position, and the test was terminated. If there was no positive response within 30 minutes, the patient was returned to the supine position, and an isoproterenol infusion of 1 mg/min (maximum infusion, 5 mg/5 min) was started (stage 3) that was increased over 5 minutes to induce a 20% increment in the resting (supine) heart rate. After titration, the patient was tilted upright again (stage 4) for 15 minutes. If a positive response occurred, the patient was returned to the supine position, and the protocol was terminated.
To exclude the confounding effects of cardiac diseases and the effects of cardiovascular risk factor conditions on myocardial contractility, we further divided the group with positive tilt test results into those with and those without co-morbid conditions. The co-morbid conditions included coronary artery disease, hypertension, diabetes mellitus, atrial fibrillation, presence of a pacemaker, and dyslipidemia. A control group from our existing database comprised healthy subjects with no history of syncope or other medical condition. These 4 groups of patients were labeled as follows: group A, positive HUTT test results with co-morbid conditions; group B, positive HUTT test results without co-morbid conditions; group C, symptomatic patients with negative HUTT test results; and group D, healthy control subjects.
All 82 patients had comprehensive resting 2-dimensional and Doppler echocardiography imaging performed within 6 months of HUTT testing. Echocardiographic studies were performed on commercially available ultrasound equipment (ACUSON Sequoia; Siemens Medical Solutions USA Inc, Mountain View, California; or Vivid 7 Dimension, GE Healthcare, General Electric, Milwaukee, Wisconsin) according to the standards recommended by the American Society of Echocardiography. Routine standard echocardiographic examination was performed, including measurement of the LV systolic and diastolic dimensions and LV ejection fraction using the modified biplane Simpson method. LV diastolic parameters were also assessed.
Echocardiographic images in digital cine-loop format (ProSolv CardioVascular Solutions, Indianapolis, Indiana) were analyzed off-line using vendor-customized 2-dimensional cardiac performance analysis software (TomTec Imaging Systems GmbH, Munich, Germany) with speckle tracking echocardiography technology for angle-independent measures of 2-dimensional strain, as described by our group in a previous study. Speckle tracking–based analysis, which can analyze 2-dimensional data from different ultrasound machines, is an extension of velocity vector imaging software previously validated with both sonomicrometry and magnetic resonance imaging.
We measured radial strain and circumferential strain from 6 segments in the short-axis view of the left ventricle at the papillary muscle level. To obtain global strain values, we also averaged longitudinal strain from the apical 4-chamber views and radial strain and circumferential strain values from the parasternal short-axis views. One observer (G.C.) who was not involved in image acquisition performed off-line analyses independent of any knowledge of the individual’s clinical presentation or other echocardiographic measures of LV function.
All continuous data are reported as mean ± SD, and categorical data are reported as number (percentage). The Kruskal-Wallis test was used for nonparametric comparisons among all 4 groups. A post hoc analysis for pairwise comparison of subgroups was performed to avoid multiple testing problems. Groups were considered to be dissimilar when the probability was p <0.05. Univariate logistic regression analysis among the clinical, echocardiographic, and strain measurements was used to determine the predictors of a positive HUTT test, whereas multivariate logistic regression analysis was used to assess independent predictors.
Interobserver and intraobserver variabilities of strain analysis from our laboratory have been presented previously. Statistical analysis was performed with commercially available statistics software (MedCalc version 11.2; MedCalc Software, Mariakerke, Belgium).
Results
Clinical baseline characteristics of the 4 study groups are summarized in Table 1 . Patients with positive HUTT test results and associated co-morbid conditions were older (age 70 ± 15 years; p <0.001) and used more medications. Use of β-blockers and calcium antagonists was not statistically different in patients with positive or negative HUTT test results.
| Characteristic | HUTT Test Positive, Co-morbid Conditions (n = 30) | HUTT Test Positive, No Co-morbid Conditions (n = 30) | HUTT Test Negative (n = 22) | Control Subjects (n = 20) | p Value |
|---|---|---|---|---|---|
| Men | 16 (53%) | 7 (23%) | 7 (32%) | 10 (50%) | 0.06 |
| Age, mean ± SD (yrs) | 70 ± 15 | 44 ± 17 | 49 ± 16 | 45 ± 17 | <0.001 |
| Heart rate, mean ± SD (beats/min) | 74 ± 13 | 72 ± 8 | 75 ± 13 | 62 ± 11 | 0.003 |
| Systolic blood pressure, mean ± SD (mm Hg) | 130 ± 23 | 117 ± 18 | 122 ± 19 | 123 ± 16 | 0.07 |
| Diastolic blood pressure, mean ± SD (mm Hg) | 72 ± 14 | 71 ± 10 | 74 ± 10 | 71 ± 12 | 0.72 |
| Presenting symptom | |||||
| Syncope | 30 (100%) | 29 (97%) | 18 (82%) | — | <0.001 |
| Palpitations | 7 (23%) | 15 (50%) | 9 (41%) | — | <0.001 |
| Chest pain | 4 (13%) | 4 (13%) | 3 (14%) | — | 0.33 |
| Shortness of breath | 5 (17%) | 3 (10%) | 3 (14%) | — | 0.24 |
| Co-morbid condition | |||||
| Coronary artery disease | 7 (23%) | — | — | — | <0.001 |
| Pacemaker | 4 (13%) | — | 1 (5%) | — | 0.05 |
| Hypertension | 23 (77%) | — | 5 (23%) | — | <0.001 |
| Diabetes mellitus | 9 (30%) | — | 1 (5%) | — | 0.001 |
| Dyslipidemia | 18 (60%) | — | 9 (41%) | — | <0.001 |
| Atrial fibrillation | 4 (13%) | — | — | — | 0.01 |
| Smoker | 1 (3%) | 4 (13%) | 2 (9%) | — | 0.21 |
| Pulmonary disease | 3 (10%) | 6 (20%) | 2 (9%) | — | 0.13 |
| Medication | |||||
| Angiotensin-converting enzyme inhibitor | 14 (47%) | — | 2 (9%) | — | <0.001 |
| Statin | 18 (60%) | 1 (3%) | 4 (18%) | — | <0.001 |
| β-blocker | 13 (43%) | 2 (7%) | 5 (23%) | — | <0.001 |
| Calcium antagonist | 6 (20%) | 1 (3%) | — | — | 0.006 |
| Digoxin | 1 (3%) | 1 (3%) | — | — | 0.67 |
| Diuretics | 6 (20%) | 1 (3%) | 2 (9%) | — | 0.04 |
| Nitrate | 2 (7%) | — | — | — | 0.16 |
The standard 2-dimensional and Doppler echocardiographic characteristics of the 4 study groups are summarized in Table 2 . The ratio of mitral early diastole to late diastole showed an increasing trend, whereas the medial ratio of early diastolic transmitral velocity to early diastolic tissue velocity showed a decreasing trend across all 4 groups; however, no statistically significant differences were found between group B and group C (i.e., study participants with positive HUTT test results and no co-morbid conditions vs study participants with negative HUTT).
| 2-Dimensional Echocardiography and Doppler Parameters | HUTT Test Positive, Co-morbid Conditions (n = 30) | HUTT Test Positive, No Co-morbid Conditions (n = 30) | HUTT Test Negative (n = 22) | Control Subjects (n = 20) | p Value |
|---|---|---|---|---|---|
| LV septum (mm) | 11 ± 1.8 ∗,† | 9 ± 1.3 | 10 ± 2 | 9.5 ± 1.3 | 0.01 |
| LV posterior wall (mm) | 10.3 ± 2 ∗ | 9 ± 1.1 † | 10 ± 2 | 10.2 ± 0.8 | 0.002 |
| LV end-diastolic dimension (mm) | 46 ± 5 † | 46 ± 5 † | 46 ± 5 † | 42 ± 5 | 0.01 |
| LV ejection fraction (%) | 61 ± 7 | 64 ± 3 | 64 ± 4 | 66 ± 4 | 0.07 |
| E/A ratio of mitral early diastole to late diastole | 1 ± 0.4 ∗,†,‡ | 1.4 ± 0.5 | 1.3 ± 0.5 | 1.7 ± 1 | <0.001 |
| Medial E/e′ | 15 ± 9 ∗,†,‡ | 8.7 ± 4 † | 8.2 ± 2 † | 5.8 ± 2.5 | <0.001 |
| Left atrial volume index ratio of early diastolic transmitral velocity to early diastolic tissue velocity (ml/m 2 ) | 30.2 ± 9 ∗ | 24 ± 6 | 26 ± 6 | 26 ± 8 | 0.04 |
| Right ventricular systolic pressure (mm Hg) | 32 ± 10 ∗ | 27 ± 5 | 27 ± 7 | — | <0.001 |
∗ Compared with the HUTT–positive group (no co-morbid conditions), p <0.05.
† Compared with the control group, p <0.05.
Table 3 summarizes the 2-dimensional strain echocardiography characteristics of the study groups. In comparison with endocardial longitudinal strain in the control group ( Figure 1 ), endocardial longitudinal strain in group A (HUTT test positive, with co-morbid conditions) and group B (HUTT test positive, with no co-morbid conditions) was markedly attenuated (p <0.05). Furthermore, compared with group C (negative HUTT test results), group A showed a significant reduction of endocardial longitudinal strain (p <0.05). Compared with group C (negative HUTT test results), group A showed a significant decrease in epicardial longitudinal strain (p <0.01). There were no differences in circumferential strain and radial strain among the groups.
| Strain Parameter | HUTT Test Positive, Co-morbid Conditions (n = 30) | HUTT Test Positive, No Co-morbid Conditions (n = 30) | HUTT Test Negative (n = 22) | Control Subjects (n = 20) | p Value |
|---|---|---|---|---|---|
| Longitudinal strain endocardial | −15 ± 5 ∗,† | −16 ± 4 † | −18 ± 4 | −19 ± 4 | 0.006 |
| Longitudinal strain epicardial | −13 ± 5 ∗ | −15 ± 4 | −17 ± 4 | −15 ± 4 | 0.02 |
| Longitudinal strain transmural gradient | −1 ± 2 † | −1 ± 2 † | −1 ± 2 † | −4 ± 3 | 0.004 |
| Circumferential strain endocardial | −19 ± 7 | −19 ± 6 | −20 ± 5 | −18 ± 8 | 0.69 |
| Circumferential strain epicardial | −6 ± 2 | −7 ± 3 | −8 ± 3 | −7 ± 4 | 0.13 |
| Circumferential strain transmural gradient | −13 ± 6 | −12 ± 4 | −12 ± 5 | −11 ± 5 | 0.65 |
| Radial strain | 17 ± 8 | 18 ± 8 | 16 ± 8 | 16 ± 8 | 0.89 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
