Background
Hypertrophic cardiomyopathy (HCM) is associated with considerable phenotypic heterogeneity. Previous studies have shown a relationship between the degree and location of hypertrophy and the prognosis of patients. The aim of this study was to compare left ventricular (LV) circumferential and longitudinal regional mechanics in patients with septal HCM and apical HCM to study the relationship between hypertrophy and function as assessed by myocardial mechanics.
Methods
Seventy-two patients with HCM (27 with apical HCM, 45 with septal HCM) were compared with 25 clinically normal and age-matched subjects. Myocardial mechanics were assessed using Velocity Vector Imaging, which extracts myocardial motion estimates from B-mode clips by tracking user-defined points and feature tracking. The Velocity Vector Imaging software generated data on global and regional systolic and diastolic longitudinal and circumferential strain, strain rate, and rotational angle velocities. One-way analysis of variance with post hoc multiple comparisons was used among the three groups.
Results
Normal subjects had relatively uniform strain and strain rates for all LV segments. Compared with the normal group, patients with septal HCM had decreased LV regional longitudinal strain rates and strain at both the basal and mid septal and lateral segments (all P < .01). Compared with patients with apical HCM, those with septal HCM had higher LV circumferential strain rates and strain at the basal and mid segments ( P < .05 to P < .01). There were significant differences in rotational velocities at the mid segments among the three groups ( P < .05 to P < .001).
Conclusions
Patients with HCM have abnormalities in myocardial mechanics that are related to the site of abnormal myocardial hypertrophy.
Hypertrophic cardiomyopathy (HCM) is a disease characterized by asymmetric left ventricular (LV) hypertrophy caused by >400 mutations in genes encoding 10 sarcomeric proteins. Phenotypically, the vast majority of patients express hypertrophy that is greatest in the interventricular septum (septal HCM). About 7% of patients with HCM demonstrate mainly apical hypertrophy (apical HCM). These two patterns of hypertrophy differ in their clinical presentation, hemodynamic characteristics, and prognoses.
The evaluation of two-dimensional (2D) myocardial mechanics has become a very sensitive echocardiographic tool to evaluate myocardial function. Regional cardiac displacement and velocity (longitudinal, circumferential, and rotation), strain, and strain rate are readily measured by 2D tissue tracking. We recently demonstrated that patients with septal HCM have decreased longitudinal strain but increased circumferential strain, with preservation of their strain vector magnitudes. LV twist was shown to be abnormal in septal HCM as mid LV rotation became clockwise rather than the normal counterclockwise direction. We have also demonstrated that patients with apical HCM have a good long-term prognoses, with a low incidence of sudden death, but have significant morbidity, which includes apical myocardial infarction and atrial fibrillation. The purpose of this study was to determine the relationship between myocardial mechanics and the pattern of hypertrophy in HCM.
Methods
Study Population
We retrospectively studied 94 adult patients (aged > 16 years) with HCM who were evaluated clinically and by echocardiography between January 2002 and December 2005 at the Toronto General Hospital. The diagnosis of HCM was established by the echocardiographic detection of septal or apical hypertrophy. Patients with septal HCM had asymmetric septal hypertrophy, with septal wall thickness ≥ 13 mm and septal wall/posterior wall thickness ratios ≥ 1.3 in the absence of a known cause for LV hypertrophy such as systemic hypertension or aortic stenosis. Patients with apical HCM had asymmetric apical hypertrophy, with apical wall thickness ≥ 15 mm and ratios of maximal apical to posterior wall thickness ≥ 1.5. All patients had normal LV systolic function by echocardiography (defined as LV ejection fraction ≥ 60%). Patient inclusion also required a Digital Imaging and Communications in Medicine echocardiographic file, with a minimum 2D frame rate of 40 frames/sec and good demarcation of the LV endocardial border.
We excluded 22 patients with HCM with associated systemic hypertension, epicardial coronary artery disease, bundle branch blocks, paced rhythms, previous interventions for the management of LV outflow tract (LVOT) obstruction, or concomitant valvular disease, leaving 72 patients for analysis (27 with apical HCM and 45 with septal HCM). The control group consisted of 25 healthy, age-matched subjects with normal clinical and echocardiographic examinations and no family histories of HCM. The study was approved by the Research Ethics Board of the Toronto General Hospital, University Health Network.
Clinical Data
Hospital records were reviewed to obtain demographic data, symptoms, New York Heart Association functional class, family history, and medications at the time of index echocardiography. Patients with HCM were followed for 34 ± 28 months (range, 6–99 months) for cardiovascular events, including atrial fibrillation, hospitalization for congestive heart failure, stroke, cardiac arrest, pacemaker or defibrillator implantation, cardiovascular procedures (septal ethanol ablation, surgical myectomy, and other cardiac surgical procedures), and death.
Doppler Echocardiographic Studies
Echocardiographic studies were interpreted by a cardiologist blinded to patients’ clinical findings. Two-dimensional, Doppler, and Doppler tissue imaging parameters were measured according to the guidelines of the American Society of Echocardiography.
Quantitation of LV Hypertrophy
Quantitative assessment of LV regional wall thickness was made from parasternal short-axis images. A total of 15 segments were analyzed, from five walls at three levels (basal, mid, and apical): the anterior interventricular septum (basal, mid, and apical), the posterior interventricular septum (basal and mid), the anterior wall (basal, mid, and apical), the lateral wall (basal, mid, and apical), the posterior wall (basal, mid, and apical), and the true apex (defined as the thickness of the apex at the junction of the apical septal and apical lateral wall segments). Maximum and average wall thickness was assessed.
Strain, Strain Rate, and Rotation Evaluation
Resting circumferential strain, strain rate, radial and rotation velocities, and angles were measured at three parasternal short-axis planes (basal, mid, and apical) in the septal and lateral walls. Longitudinal septal and lateral wall strain and strain rates were measured from the apical four-chamber view, at the basal, mid, and apical levels. Measurements were performed using Velocity Vector Imaging (VVI) version 1.0 (Siemens Medical Systems, Mountain View, CA) from archived 2D echocardiographic studies as previously described. The VVI version used did not provide positive radial strain (thickening) analysis.
Postprocessing Calculations
Circumferential regional LV mechanics were analyzed at the three (basal, mid, and apical) short-axis views for the septal, anterior, lateral, and inferior walls. In the apical four-chamber view, regional LV longitudinal mechanics were analyzed for six segments (basal, mid, and apical for the septal and lateral walls). Averaged myocardial rotation angles were used to calculate LV twist. To standardize the temporal evaluation of twist regardless of heart rate, all times were normalized and presented as percentages of the cycle length. Twist was defined as the maximal instantaneous basal-to-apical angle difference.
Reproducibility
Five normal subjects and five patients with HCM from the study cohort were randomly selected, and their echocardiographic studies were analyzed by another investigator and by the same investigator 1 month later. The linear correlation and standard deviation of percentage difference between intraobserver and interobserver measurements were calculated for the LV mechanic variables.
Statistical Analysis
Summary statistics are expressed as mean ± SD. One-way analysis of variance was used with least square difference post hoc multiple comparisons to compare LV regional mechanics among the normal, apical HCM, and septal HCM groups and among differing degrees of LV hypertrophy. Categorical data were compared using χ 2 or Fisher’s exact tests as appropriate. The correlations between strain parameters and wall thickness were determined by Pearson’s correlation coefficients. A two-tailed P value < .05 was considered statistically significant. SPSS version 11.0 (SPSS, Inc., Chicago, IL) was used for statistical analysis.
Results
Distribution of LV Hypertrophy
LV wall thickness at 15 measurement sites were taken from patients with septal and apical HCM and are summarized in Figure 1 . The pattern of increased thickness differed significantly between the two HCM groups, as expected. In patients with septal HCM, all septal segments were abnormal, with the posterior and lateral walls relatively spared at the basal and mid levels and the septum primarily involved at the basal and mid levels. In patients with apical HCM, average values for all apical wall segments were all increased.
Clinical Characteristics
The mean ages were similar in the three groups studied (47.9 ± 12.4, 54.2 ± 16.4, and 49.5 ± 17.4 years for 25 normal controls, 27 patients with apical HCM, and 45 patients with septal HCM, respectively). Table 1 compares the clinical characteristics of patients with apical HCM and those with septal HCM at the time of index echocardiography and during follow-up. Patients with septal HCM had higher New York Heart Association functional classes and were more symptomatic, primarily with a higher incidence of dyspnea and presyncope. Patients with septal HCM also differed from those with apical HCM in terms of a significantly higher prevalence of LVOT obstruction, significant mitral regurgitation, and procedures to alleviate LVOT obstruction. Arrhythmic complications and related device implantation procedures did not significantly differ between the groups, but there was a trend toward higher rates in the patients with septal HCM.
| Patients with apical HCM | Patients with septal HCM | |
|---|---|---|
| Clinical data | (n = 27) | (n = 45) |
| Age (years) | 54.2 ± 16.4 | 49.5 ± 17.4 |
| Men | 81% | 60% |
| Presentation | ||
| NYHA functional class | 1.5 ± 0.7 | 2.2 ± 0.8 † |
| NYHA functional class ≥ III | 12% | 42% ∗ |
| Dyspnea | 44% | 74% ∗ |
| Chest pain | 40% | 64% |
| Presyncope | 20% | 62% † |
| Syncope | 20% | 26% |
| Atrial fibrillation | 20% | 10% |
| Echocardiography | ||
| MR grade (grading, 1–4) | 1.4 ± 0.6 | 2.4 ± 1.2 ∗ |
| Moderate or greater MR | 0 | 23% ∗ |
| LVOT obstruction (pressure gradient ≥ 30 mm Hg) | 0 | 63% ∗ |
| Resting LVOT gradient (mm Hg) | 6.8 ± 1.4 | 39.8 ± 35.4 ∗ |
| Clinical outcomes | ||
| VT/VF | 0 | 5% |
| Stroke | 0 | 2% |
| Hospitalization for CHF | 7% | 7% |
| Pacemaker/ICD implantation | 8% | 17% |
| Procedures for LVOT obstruction (SEA/myectomy) | 0 | 24% † (6/5) |
Conventional 2D and Doppler Echocardiography
LV systolic dimensions were normal in patients with apical HCM and smaller than normal in those with septal HCM ( Table 2 ). Diastolic parameters differed among the analyzed groups. Patients with apical HCM had lower early diastolic mitral inflow and annular velocities, longer isovolumic relaxation times, and increased left atrial volumes indexed to body surface area compared with normal controls. Patients with septal HCM had normal mitral early diastolic flow velocities, low mitral annular velocities, and prolonged isovolumic relaxation times compared with normal controls. Mitral early diastolic deceleration times were longer, and indexed left atrial volumes and calculated E/E′ ratios were higher, compared with normal controls and patients with apical HCM. Thus, diastolic dysfunction was evident in both patterns of HCM.
| Normal controls | Patients with apical HCM | Patients with septal HCM | |
|---|---|---|---|
| Variable | (n = 25) | (n = 27) | (n = 45) |
| LVd (cm) | 4.51 ± 0.39 ∗ | 4.87 ± 0.50 ‡ | 4.35 ± 0.57 |
| LVs (cm) | 2.87 ± 0.40 | 2.92 ± 0.52 ‡ | 2.38 ± 0.59 † |
| MVE (m/sec) | 0.76 ± 0.13 ∗ | 0.62 ± 0.15 ‡ | 0.78 ± 0.26 |
| MVA (m/sec) | 0.61 ± 0.15 | 0.51 ± 0.17 ‡ | 0.71 ± 0.33 |
| DT (msec) | 203 ± 40 | 201 ± 58 ‡ | 241 ± 70 † |
| IVRT (msec) | 74 ± 16 ∗ | 87 ± 29 | 91 ± 20 † |
| PVS (m/sec) | 0.55 ± 0.10 | 0.50 ± 0.13 | 0.55 ± 0.12 |
| PVD (m/sec) | 0.46 ± 0.13 | 0.44 ± 0.18 | 0.42 ± 0.12 |
| E′ (cm/sec) | 13.0 ± 4.2 ∗ | 8.6 ± 2.5 | 7.3 ± 3.6 † |
| E/E′ | 6.1 ± 1.9 | 6.1 ± 2.2 | 13.9 ± 11.1 † |
| LAVI (mL/m 2 ) | 24.4 ± 6.9 ∗ | 35.4 ± 8.3 | 40.2 ± 12.6 † |
∗ P < 0.05, patients with apical HCM vs normal controls.
† P < .05, patients with septal HCM vs normal controls.
‡ P < .05, patients with apical HCM vs patients with septal HCM.
Regional Myocardial Mechanics
Longitudinal Strain
Long-axis shortening (longitudinal strain; Figure 2 A) was uniform in the three levels examined in the normal control group (−21 ± 4%). In the two groups of patients with HCM, longitudinal strain varied with the pattern of regional hypertrophy. Compared with normal controls, patients with apical HCM had decreased longitudinal strain in the hypertrophied mid and apical segments, while nonthickened basal segments demonstrated normal longitudinal strain. In patients with septal HCM, longitudinal strain was lower than in normal controls in all segments except the apical lateral segment. In these patients, the extent of the decrease in longitudinal strain followed the extent of segmental hypertrophy ( Figures 2 and 3 ). Thus, regional changes reflected the pattern of hypertrophy in both patient groups, with reductions of longitudinal strain of the mid segments common to both groups.
Longitudinal Strain Rate
Because the rate of long-axis shortening (strain rate S) paralleled the changes in strain, it did not introduce new information and is not discussed. Long-axis early diastolic stretching rate (longitudinal strain rate E) was reduced in patients with apical HCM at the mid and apical lateral segments compared with normal controls, whereas in patients with septal HCM, it was reduced at the basal and mid levels. Interestingly, there were no differences between apical HCM and septal HCM in strain rate E. The late diastolic longitudinal strain rate (strain rate A) was highly variable and did not differ between normal controls and the two HCM groups.
Circumferential Strain
Apical-to-basal circumferential strain gradients were present in normal controls and in both HCM groups. Circumferential strain showed similar regional variability in HCM as longitudinal strain, but in the opposite direction. Instead of being lower in thickened segments (less negative), it was found to be higher (more negative). Compared with normal controls, patients with septal HCM had increased circumferential strain at the basal and mid septal segments and the mid anterior segment. Compared with patients with apical HCM, those with septal HCM had increased circumferential strain at the basal and mid septum and mid anterior segment ( Figure 2 ).
Circumferential Strain Rate
Systolic circumferential strain rate demonstrated changes similar to circumferential strain. Circumferential diastolic strain rate E and A did not differ among the three groups ( P > .05).
Correlations Between LV Regional Thickness and Mechanics
There were 393 pairs of wall thickness and longitudinal strain and early diastolic strain rate (L-SR-E), 25 pairs from the normal group (basal septal segments) and 368 pairs from two HCM groups (septal and lateral walls at three levels), available for correlation assessment ( Figure 4 ). Longitudinal strain and diastolic strain rate varied according to the degree of hypertrophy: in segments with wall thicknesses ≤ 15 mm, the average strain was −18.2 ± 5.5% and L-SR-E was 0.94 ± 0.43 s −1 ; in segments with thicknesses of 16 to 20 mm, strain was −15.5 ± 5.1% and L-SR-E was 0.76 ± 0.40 s −1 ; and in segments with thicknesses > 20 mm, strain was −13.8 ± 5.3% and L-SR-E was 0.72 ± 0.32 s −1 . There were significant differences in strain among the three groups when classified by degree of wall thickness ( P < .00001), representing a stepwise decrease in longitudinal strain with increased LV wall thickness. There was also a difference in L-SR-E between wall thickness ≤ 15 mm and the other two groups ( P = .0002). However, no difference was found between thickness of 16 to 20 mm and thickness > 20 mm. Regional thickness was found to correlate with the longitudinal strain and strain rate E ( r = 0.50 and r = 0.53, respectively, P < .0001; Figures 4 A and 4 B).
Rotation
Rotation in the normal subjects was clockwise (as imaged from the apex) at the basal level and anticlockwise at the mid and apical levels. The rotation velocities and directions of patients with HCM were similar to those of normal subjects at the basal and apical levels, yielding normal LV maximal twist (maximal instantaneous basal-to-apical angle difference). However, regardless of the pattern of hypertrophy in patients with HCM, rotational velocities at the mid segment followed the basal direction (clockwise), opposite to the normal anticlockwise direction of the normal group ( Figure 5 ). Twist and twist time (time to peak basal-to-apical angle difference, standardized to cycle length) was similar in the three groups (45 ± 14% of cycle length). The rotational velocity of the mid segment in patients with apical HCM still followed that of the basal direction, but the absolute value was significantly decreased compared with that at the base ( Figure 5 ).
