Refractory progressive heart failure (HF) is becoming the predominant cause of mortality in nonobstructive hypertrophic cardiomyopathy (HC). To anticipate development of this important and often unpredictable clinical course, we investigated whether left ventricular diastolic dysfunction, assessed by echocardiographic Doppler parameters, could identify a subset of patients with HC without obstruction at rest who would experience progression of HF. Diastolic function parameters, assessed by Doppler tissue imaging (DTI), mitral inflow, and pulmonary venous flow were measured in 274 consecutive adult patients with HC evaluated from 2003 to 2007. DTI and other diastolic and clinical/demographic parameters were measured against the composite end point of HF/death, heart transplantation, or progression to advanced New York Heart Association functional class III/IV symptoms and sudden death (SD)/implantable defibrillator (ICD) interventions. HF end points were reached in 19 of 274 patients (7%) over a follow-up period of 4.0 ± 2.3 years. Variables significantly associated with HF outcome by univariate analysis included male gender, initial New York Heart Association class II, lower ejection fraction, and reduced septal and lateral e′ mitral annular tissue velocities. Multivariable analysis showed only a reduced lateral e′ mitral annular tissue velocity to be independently associated with the composite HF end points (HR 0.77; 95% CI 0.65 to 0.91; p = 0.003). In addition, estimated pulmonary arterial systolic pressure and extensive late gadolinium enhancement by magnetic resonance were also associated with HF outcome (p = 0.04 and p <0.001, respectively). No Doppler (or clinical) variable was associated with SD/appropriate ICD interventions. In conclusion, in HC without outflow obstruction at rest, diastolic dysfunction, evidenced by DTI-reduced lateral e′ mitral annular tissue velocity, was associated with adverse long-term HF outcome but was unrelated to SD. This echocardiographic marker provides a potential noninvasive strategy for anticipating progressive HF in this HC patient group.
Hypertrophic cardiomyopathy (HC) is a common genetic heart disease with a risk for sudden death (SD) in the young and heart failure (HF) symptoms and disability over a wide range of ages. Although relatively infrequent, SD has been regarded as the most visible and devastating consequence of HC, with several established risk markers predictive of these events. However, HF symptoms are a more common complication of HC occurring in about 50% patients, largely the consequence of left ventricular (LV) outflow tract gradients. Although LV diastolic dysfunction is a hallmark of HC, evident in most patients, it is incompletely resolved as to whether diastolic function parameters reliably identify patients without obstruction who are likely to develop HF symptoms. Therefore, because additional markers are required to anticipate development and progression of HF, we determined in a large cohort of patients with HC (without subaortic obstruction at rest) whether Doppler tissue imaging (DTI), or other Doppler echocardiographic measures are associated with the progression of HF during the clinical course of HC.
Methods
We evaluated 404 consecutive adult patients presenting to the Hypertrophic Cardiomyopathy Center of the Minneapolis Heart Institute Foundation, from 2003 to 2007. Standard transthoracic 2-dimensional and Doppler echocardiographic assessments (including DTI) were obtained during the initial outpatient visit. Of these 404 patients, 130 were excluded from the cohort on the basis of LV outflow obstruction (n = 87) with peak instantaneous gradients ≥30 mm Hg, New York Heart Association (NYHA) class III to IV symptoms at study entry (n = 36), or technically inadequate echocardiograms (n = 7). Therefore, the final study group comprised 274 patients with HC without obstruction at rest who had no or only mild symptoms (NYHA classes I to II) at study entry.
HC was defined by a hypertrophied and nondilated LV (wall thickness ≥13 mm) present at some point during the patient’s clinical course, and in the absence of another cardiac or systemic disease capable of producing a similar magnitude of hypertrophy. Patients were 42 ± 18 years at study entry and initial echocardiographic study. No study patient was under consideration for surgical septal myectomy or alcohol ablation.
The study protocol was approved by the Schulman Associates Institutional Review Board. All investigators had full access to, and take responsibility for, the integrity of the data and have agreed to the manuscript as written.
Two-dimensional and Doppler echocardiographic studies were performed with an ACUSON Sequoia System at study entry. For DTI, a sample volume was positioned within 1 cm of septal and lateral insertion sites of mitral leaflets at the annulus, as visualized in apical views. Spectral recordings were obtained at sweep speed 50 to 100 mm/s at end-expiration and measured in early diastole (e′ [cm/s]).
Pulse-wave Doppler was performed in the apical 4-chamber view using color-flow imaging for optimal alignment of the ultrasound beam. The sample volume was placed between mitral leaflet tips during diastole to record velocity profiles (early [E] and late [A] waves [cm/s]). Spectral velocity waveforms (systolic [S] and diastolic [D] waves [cm/s]) were recorded by placement of the sample volume in the right upper pulmonary vein as visualized in the apical 4-chamber view. Peak pulmonary artery systolic pressure (in mm Hg) was estimated with continuous-wave Doppler by the tricuspid regurgitant jet velocity squared (x4) and summed with right atrial pressure (inferred from inferior vena cava dimension with respect to respiratory variation).
A single observer, AK, measured diastolic function parameters: transmitral pulse-wave, pulmonary vein and tricuspid regurgitant jet velocities, DTI parameters, and standard measurements of LV and left atrial chamber dimensions, blinded to knowledge of clinical findings and course. Diastolic parameters were measured on 3 consecutive complexes in sinus rhythm and averaged. In those patients with a history of paroxysmal atrial fibrillation, all measurements were made in sinus rhythm.
Cardiovascular magnetic resonance imaging was performed in 267 consecutive patients, with a 1.5-T scanner (Philips, Best, The Netherlands; or Siemens, Erlangen, Germany) using steady-state free-precession breath-hold cines in 3 long-axis planes and sequential short-axis slices from the atrioventricular ring to the apex. Late gadolinium enhancement images were acquired 10 to 20 minutes after intravenous administration of 0.2 mmol/kg gadolinium-diethylenetriamine pentaacetic acid with breath-hold 2-dimensional segmented inversion-recovery sequence or phase-sensitive inversion-recovery sequences in identical planes as in cine images. Late gadolinium enhancement quantification was performed by manually adjusting grayscale threshold visually, which was summed and expressed as a proportion of total LV myocardium.
To test interobserver agreement, 2 readers blinded to clinical findings (AK and KMH) independently measured septal and lateral e′ mitral annular tissue velocities, and E- and A-wave amplitudes, accessing every 20th echocardiogram from the study cohort with patients arranged in alphabetical order. Correlation coefficients for interobserver variability were r 2 = 0.97 for septal e′ mitral annular tissue velocity, r 2 = 0.97 for lateral e′ mitral annular tissue velocity, r 2 = 0.83 for E-wave amplitude, and r 2 = 0.78 for A-wave amplitude.
Descriptive statistics are displayed as mean and SD for continuous variables and number and percentage for categorical variables. These variables were assessed for significance using Cox proportional hazard models. Doppler echocardiographic parameters, defined in advance, were tested against a composite end point of HC-related HF death, heart transplant, and progression to NYHA class III or IV. All relevant clinical or Doppler echocardiographic variables were tested for univariate associations. Variables with p ≤0.05 on univariate analysis were entered into a stepwise multivariable Cox proportional hazard model: gender, initial NYHA class, initial ejection fraction, transmitral E and A velocities, and DTI—septal and lateral e′ mitral annular tissue velocities. Proportional hazards were tested graphically before proceeding. p Values are 2 sided where appropriate. All analyses were performed using Stata 11.2 (StataCorp LP, College Station, Texas).
Results
Clinical and demographic characteristics of the study population are summarized in Table 1 . Age at first evaluation was 41 ± 18 years; 207 patients (76%) were men. Maximum LV wall thickness was 22.1 ± 6.3 mm (range 13 to 44). Ejection fraction was 65 ± 6%, including 4 patients <50%. Most commonly, by convention, patients were administered β blockers and/or verapamil during their clinical course for symptom control ( Table 1 ).
| Variable No. (%) or mean (SD) | All Patients – Study Entry (n=274) | HC-HF Death/ Transplant/ NYHA Class III-IV (n=19) | Without HC-HF end-point (n=255) | P-Value | SD/ICD | P-Value | |
|---|---|---|---|---|---|---|---|
| Interventions (n=13) | No Event (n=261) | ||||||
| Age (Years) | 41.2 ± 18.1 | 43.7 ± 17.9 | 41.0 ± 18.1 | 0.56 | 39.3 ± 18.5 | 41.3 ± 18.1 | 0.61 |
| Female gender | 67 (24%) | 9 (47%) | 58 (23%) | 0.02 | 6 (46%) | 61 (23%) | 0.078 |
| Maximum LV thickness (mm) | 22.1 ± 6.3 | 21.9 ± 3.4 | 22.1 ± 6.5 | 0.84 | 25.3 ± 8.2 | 21.9 ± 6.2 | 0.046 |
| Initial NYHA class | |||||||
| I | 209 (76%) | 9 (47%) | 200 (78%) | 0.004 | 10 (77%) | 199 (76%) | 0.97 |
| II | 65 (24%) | 10 (53%) | 55 (22%) | 3 (23%) | 62 (24%) | ||
| History of paroxysmal atrial fibrillation | 68 (25%) | 8 (42%) | 60 (24%) | 0.11 | 2 (15%) | 52 (20%) | 0.68 |
| MR – Severe | 2 (1%) | 0 | 2 (1%) | 1.0 | 0 | 2 (1%) | 1.00 |
| Systolic pulmonary arterial pressure (estimated) | 26.1 ± 9.5 | 32.1 ± 10.6 | 25.6 ± 9.2 | 0.042 | 23.4 ± 5.9 | 26.3 ± 9.6 | 0.47 |
| EF (echo) | 65.2 ± 6.7 | 61.0 ± 11.8 | 65.5 ± 6.1 | 0.002 | 65.8 ± 8.1 | 65.2 ± 6.6 | 0.74 |
| PA pressure grade ∗ | |||||||
| Normal | 78 (70%) | 5 (50%) | 73 (72%) | 6 (86%) | 72 (69%) | ||
| Mild increase | 31 (28%) | 4 (40%) | 27 (26%) | 0.093 | 1 (14%) | 30 (29%) | 0.58 |
| Moderate increase | 3 (3%) | 1 (10%) | 2 (2%) | 0 | 3 (3%) | ||
| LA diameter (mm) | 41 ± 7.5 | 43.5 ± 8.9 | 40.8 ± 7.4 | 0.13 | 37.7 ± 9.4 | 41.2 ± 7.4 | 0.10 |
| LGE present | 80 (41%) | 9 (53%) | 71 (40%) | 0.27 | 7 (78%) | 73 (39%) | 0.044 |
| LGE amount (%LV) † | 6.4 ± 9.3 | 16.7 ± 18.8 | 5.1 ± 6.5 | <0.001 | 12.8 ± 12.9 | 5.8 ± 8.7 | 0.062 |
| % LGE ≥ 15% LV | 9/80 (11%) | 4/9 (44%) | 5/71 (7%) | 0.001 | 2/7 (29%) | 7/73 (10%) | 0.12 |
| Risk Factors | |||||||
| NSVT | 49 (18%) | 4 (21%) | 45 (18%) | 0.70 | 4 (31%) | 45 (17%) | 0.25 |
| Unexplained syncope | 48 (18%) | 6 (32%) | 42 (16%) | 0.13 | 3 (23%) | 45 (17%) | 0.65 |
| Family history SD | 65 (24%) | 3 (16%) | 62 (24%) | 0.35 | 3 (23%) | 62 (24%) | 0.96 |
| LV thickness ≥ 30 mm | 37 (13%) | 1 (5%) | 36 (14%) | 0.32 | 3 (23%) | 34 (13%) | 0.27 |
| Medications-Last Evaluation | |||||||
| beta-blockers | 139 (51%) | 10 (53%) | 129 (51%) | 0.93 | 7 (54%) | 132 (51%) | 0.83 |
| calcium channel antagonists | 60 (22%) | 3 (16%) | 57 (22%) | 0.46 | 3 (23%) | 57 (22%) | 0.95 |
| amiodarone | 8 (3%) | 0 | 8 (3%) | 1.00 | 2 (16%) | 6 (2%) | 0.006 |
| disopyramide | 6 (2%) | 0 | 6 (2%) | 1.00 | 0 | 6 (2%) | 1.00 |
∗ Peak systolic pulmonary artery pressure grades: normal <35 mm Hg; mild 35 to 49 mm Hg; moderate 50 to 60 mm Hg; severe >60 mm Hg.
During a follow-up period of 4.0 ± 2.3 years, the composite end point of HC-related HF death (n = 4) or heart transplantation (n = 4) or progression to NYHA class III to IV symptoms (n = 11) was reached in 19 patients (7%), but in none of the other 255 study patients. Of the 274 patients, 13 died suddenly (n = 2) or experienced appropriate implantable cardioverter-defibrillator (ICD) interventions for ventricular tachycardia/VF (n = 11).
In univariate analysis, 5 variables were significant with respect to the composite HF end-point: male gender, initial NYHA class II, reduced ejection fraction, reduced e′ septal velocity, and also reduced lateral e′ mitral annular tissue velocity which was significantly lower in patients who reached the composite HF end point ( Tables 2 and 3 ): 9.1 (3.4) cm/s versus 12.7 (4.3) cm/s in patients who did not achieve this end point; p <0.001 ( Figure 1 ; Table 2 ). Furthermore, multivariable regression analysis showed reduced lateral e′ mitral annular tissue velocity to be independently associated with HC-related HF outcome (HR 0.77; 95% CI: 0.65 to 0.91; p = 0.003; Table 3 ).
| DTI Diastolic Function Parameter | HC – HF Death/ Transplant/NYHA Class III/IV (n = 19) | Without HC – HF Death/ Transplant/NYHA Class III/IV (n = 255) | P-Value |
|---|---|---|---|
| E/e′ septal | 10.5 ± 4.8 | 9.6 ± 4.1 | 0.38 |
| E/e′ lateral | 7.9 ± 3.3 | 6.9 ± 3.2 | 0.13 |
| E/e′ (mean of septal and lateral) | 8.3 ± 3.3 | 7.9 ± 3.4 | 0.49 |
| e′ septal (cm/s) | 7.9 ± 3.1 | 9.1 ± 2.9 | 0.044 |
| e′ lateral (cm/s) | 9.1 ± 3.4 | 12.7 ± 4.3 | 0.002 |
| e′ (mean of septal and lateral) (cm/s) | 8.6 ± 3.2 | 10.9 ± 3.3 | 0.007 |
There was no statistically significant difference in septal E/e′ or lateral E/e′ ratios between these 2 groups ( Table 2 ). Also, DTI measures did not differ between patients with or without SD/ICD interventions ( Table 4 ).
