In hypertrophic cardiomyopathy (HCM), the clinical significance attributable to the broad range of left ventricular (LV) systolic function, assessed as the ejection fraction (EF), is incompletely resolved. We evaluated the EF using cardiovascular magnetic resonance (CMR) imaging in a large cohort of patients with HCM with respect to the clinical status and evidence of left ventricular remodeling with late gadolinium enhancement (LGE). CMR imaging was performed in 310 consecutive patients, aged 42 ± 17 years. The EF in patients with HCM was 71 ± 10% (range 28% to 89%), exceeding that of 606 healthy controls without cardiovascular disease (66 ± 5%, p <0.001). LGE reflecting LV remodeling showed an independent, inverse relation to the EF (B-0.69, 95% confidence interval −0.86 to −0.52; p <0.001) and was greatest in patients with an EF <50%, in whom it constituted a median value of 29% of the LV volume (interquartile range 16% to 40%). However, the substantial subgroup with low-normal EF values of 50% to 65% (n = 45; 15% of the whole cohort), who were mostly asymptomatic or mildly symptomatic (37 or 82% with New York Heart Association functional class I to II), showed substantial LGE (median 5% of LV volume, interquartile range 2% to 10%). This overlapped with the subgroup with systolic dysfunction and significantly exceeded that of patients with an EF of 66% to 75% and >75% (median 2% of the LV volume, interquartile range 1.5% to 4%; p <0.01). In conclusion, in a large cohort of patients with HCM, a subset of patients with low-normal EF values (50% to 65%) was identified by contrast-enhanced CMR imaging as having substantial degrees of LGE, suggesting a transition phase, potentially heralding advanced LV remodeling and systolic dysfunction, with implications for clinical surveillance and management.
Hypertrophic cardiomyopathy (HCM) is widely regarded as a disease predominantly associated with hyperdynamic left ventricular (LV) systolic function, although the disease of a few patients is known to evolve into overt systolic dysfunction and the so-called end-stage phase. To date, the potential clinical significance attributable to the wide range in measured ejection fractions (EFs) is incompletely resolved, in part because of the inherent limitations of 2-dimensional echocardiography in the accurate quantification of the LV volume in this disease. Cardiovascular magnetic resonance (CMR) imaging, with contrast enhancement, because of its high-resolution volumetric reconstruction of the LV chamber, affords a highly accurate and reproducible quantitative assessment of LV size and systolic function. It also provides in vivo contrast visualization of late gadolinium enhancement (LGE), generally considered indicative of myocardial fibrosis. Therefore, in the present study, we evaluated LV systolic function by CMR imaging in a large cohort of patients with HCM, to characterize the early stages of disease progression preceding systolic dysfunction.
Methods
The study population included 310 patients with HCM consecutively referred for CMR imaging from 2001 to 2008 at centers in Minneapolis and Boston ( Table 1 ). CMR imaging was routinely offered to all patients with HCM evaluated at our institutions during the study period for the purpose of defining the extent and distribution of LV hypertrophy, LV volumes, LV mass, and LGE. Those with specific contraindications, such as implanted cardioverter-defibrillators or pacemakers, metallic fragments, known claustrophobia, renal insufficiency, and pregnant or lactating women were excluded. The diagnosis of HCM was determined from CMR imaging and 2-dimensional echocardiographic evidence of a hypertrophied left ventricle (maximum wall thickness ≥15 mm) with a normal or small cavity size (defined by an end-diastolic volume index <75 ml/m 2 ), in the absence of another cardiac or systemic disease that could produce the magnitude of hypertrophy evident, at some point during the patient’s clinical course. None of our patients had significant coronary artery disease (defined as >50% stenosis in one major artery), as ascertained by specific clinical and/or CMR evidence. First, no patient had experienced an acute coronary event associated with increased cardiac enzymes or Q waves on the electrocardiogram. Second, when LGE was present in a subendocardial or transmural distribution within a single coronary artery vascular territory, hemodynamically significant coronary artery disease was excluded by arteriography or computed tomography angiography. Patients with previous cardiac surgery or percutaneous alcohol septal ablation were excluded from the study. The respective internal review board or research ethics committees of each participating institution approved the study protocol, and all subjects provided written inform consent. Selected data from subsets in this patient cohort have been reported as a part of other analyses.
| Clinical/Demographic Data | Overall | Ejection Fraction | p Value | |||
|---|---|---|---|---|---|---|
| <50% ⁎ | 50–65% | 66–75% | >75% | |||
| Patients | 310 | 15 | 45 | 144 | 106 | |
| Men | 218 (70%) | 9 (60%) | 33 (73%) | 110 (76%) | 66 (62%) | 0.08 |
| Age at study entry (years) | 42 ± 17 | 43 ± 17 | 40 ± 17 | 41 ± 18 | 46 ± 17 | 0.12 |
| Age at diagnosis (years) | 37 ± 17 | 30 ± 17 | 35 ± 15 | 36 ± 18 | 41 ± 16 | 0.04 |
| New York Heart Association functional class at study entry | 1.3 ± 0.6 | 1.5 ± 05 | 1.4 ± 0.6 | 1.2 ± 0.5 | 1.4 ± 0.6 | 0.08 |
| I | 221 (71%) | 4 (27%) | 29 (64%) | 117 (81%) | 71 (67%) | |
| II | 43 (14%) | 8 (53%) | 8 (18%) | 8 (6%) | 19 (18%) | |
| III | 46 (15%) | 3 (20%) | 8 (18%) | 19 (13%) | 16 (15%) | |
| Angina pectoris | 109 (35%) | 3 (20%) | 17 (38%) | 51 (35%) | 38 (36%) | 0.64 |
| Syncope | 68 (22%) | 4 (27%) | 10 (22%) | 30 (21%) | 24 (23%) | 0.95 |
| Atrial fibrillation | 39 (13%) | 6 (40%) † | 5 (11%) | 19 (13%) | 9 (8%) | 0.019 |
| Medical treatment | 220 (71%) | 13 (87%) | 24 (53%) | 92 (64%) | 91 (86%) ‡ § | <0.01 |
| β Blockers | 179 (58%) | 12 (80%) | 19 (42%) | 76 (53%) | 72 (68%) ‡ | 0.004 |
| Verapamil | 65 (21%) | 2 (13%) | 8 (18%) | 26 (18%) | 29 (27%) | 0.24 |
| Amiodarone | 6 (2%) | 2 (13%) ¶ | 0 | 3 (2%) | 1 (1%) | 0.008 |
| Disopyramide | 7 (2%) | 0 | 0 | 6 (4%) | 1 (1%) | 0.20 |
| Diuretics | 38 (12%) | 5 (33%) § | 3 (7%) | 11 (8%) | 19 (18%) | 0.004 |
| Angiotensin-converting enzyme inhibitors | 39 (13%) | 6 (40%) † | 5 (11%) | 14 (10%) | 14 (13%) | 0.01 |
| Systemic hypertension | 72 (23%) | 3 (20%) | 10 (22%) | 30 (21%) | 29 (27%) | 0.66 |
| Implantable cardioverter-defibrillator | 59 (19%) | 7 (47%) § ¶ | 11 (24%) | 23 (16%) | 18 (17%) | 0.03 |
| Echocardiography | ||||||
| Left atrium (mm) | 42 ± 8 | 45 ± 10 | 44 ± 7 | 41 ± 8 | 42 ± 7 | 0.22 |
| Maximum left ventricular wall thickness (mm) | 21 ± 6 | 21 ± 7 | 21 ± 7 | 21 ± 6 | 21 ± 5 | 0.91 |
| Left ventricular end-diastolic diameter (mm) | 44 ± 7 | 48 ± 11 ¶ | 47 ± 6 ¶ | 44 ± 6 | 43 ± 7 | 0.002 |
| Left ventricular outflow tract obstruction at rest (≥30 mm Hg) | 61 (20%) | 0 | 7 (16%) ¶ | 24 (17%) | 30 (28%) | 0.03 |
| Moderate-to-severe mitral regurgitation | 26 (8%) | 1 (7%) | 3 (7%) | 9 (6%) | 13 (12%) | 0.36 |
† p <0.05 versus all other groups;
A reference population of 606 healthy adult participants in the Framingham Heart Study Offspring Cohort (239 men and 367 women) without evidence of clinical cardiovascular disease underwent CMR imaging, using a scanning protocol similar to that reported in the present study for patients with HCM. The mean age was 61 ± 8 years for both men and women. The body surface area was 2.0 ± 0.2 m 2 for men and 1.7 ± 0.2 m 2 for women.
Comprehensive 2-dimensional and Doppler echocardiographic studies were performed for each patient using commercially available instruments. LV hypertrophy was assessed by 2-dimensional echocardiography, and the site and extent of the maximum wall thickness were identified. The peak instantaneous LV outflow gradient, resulting from mitral valve systolic anterior motion and mitral–septal contact, was estimated with continuous wave Doppler under basal conditions. The left atrial dimension was measured at end-systole in the anteroposterior linear diameter from the parasternal long-axis view.
CMR imaging was performed (Philips Gyroscan ACS-NT 1.5 T, Best, The Netherlands; and Siemens Sonata 1.5 T, Erlangen, Germany) using steady-state, free precession breath-hold cines in 3 long-axis planes and contiguous 10-mm (no gap) or 8-mm (2-mm gap) short-axis slices from the atrioventricular ring to the apex.
All measurements on the CMR studies in the patients with HCM and controls were performed by a centralized core laboratory (PERFUSE Angiographic Core Laboratory and Data Coordinating Center, Harvard Medical School, Boston, Massachusetts), previously used in other studies. The LV volumes, ejection fraction, mass, and wall thickness were analyzed using a commercial workstation (MASS, version 6.1.6, Medis, Best, The Netherlands). The endocardial and epicardial borders of the left ventricle were manually measured using planimetry by an experienced observer (CJH) on successive short-axis cine images at end-diastole, with only the endocardial border measured using planimetry on the end-systolic frame. The LV volume and mass were derived by summation of disks, with the mass calculated by multiplying the myocardial muscle volume by 1.05 g/cm 3 . The LV EF was calculated by dividing the LV stroke volume by the end-diastolic volume. The maximum LV wall thicknesses were taken as the greatest dimension determined automatically by the MASS software at any site within the LV wall. The anatomic parameters were normalized to the body surface area.
LGE images were acquired 10 to 15 minutes after intravenous administration of 0.2 mmol/kg gadolinium-diethylene triamine pentaacetic acid (Magnevist, Schering) with a breath-held segmented inversion-recovery sequence, acquired in the same orientations as the cine images. An inversion time scout was used initially to find the optimal inversion time to null the normal myocardium (typically 240 to 300 ms).
To ascertain the presence of LGE, all tomographic short-axis LV slices from base to apex were inspected visually to identify an area of completely nulled myocardium. The mean signal intensity (and SD) of the normal myocardium was calculated, and a threshold ≥6 SD exceeding the mean was used to define areas of LGE. The choice of 6 SD was determined from our experience that semiautomated LGE-CMR gray-scale thresholding using ≥6 SD greater than the mean of visually normal, remote myocardium was the most reliable method for assessing the extent of LGE in the LV myocardium of patients with HCM.
The total LGE volume (expressed in grams) was calculated by summing the planimetered areas of LGE present on each short-axis slice and multiplying it by the slice thickness (10 mm). It was expressed as a proportion of the total LV myocardial volume (percentage of LGE).
The unpaired Student t test or one-way analysis of variance followed by Bonferroni’s post hoc test were used for the comparison of normally distributed data. The Kruskal-Wallis H test was used to compare the extent of LGE (which was not normally distributed) across different EF subgroups. The chi-square test or Fisher’s exact test was used to compare noncontinuous variables, expressed as proportions. Multivariate linear regression analysis was performed to assess the relation of the EF to several clinical variables and LGE. p Values are 2-sided and considered significant at <0.05. The calculations were performed using the Statistical Package for Social Sciences, version 12.0, software (SPSS, Chicago, Illinois).
Results
The EF in the 310 patients with HCM was significantly greater than in the reference healthy control population (71 ± 10% vs 66 ± 5%; p <0.001). The EF was <50% (overt systolic dysfunction) in 15 (5%), 50% to 65% in 45 (15%), 66% to 75% in 144 (46%), and >75% in 106 (34%). Patients with an EF <50% and an EF of 50% to 65% had similar transverse LV end-diastolic and left atrial dimensions, exceeding those of patients with an EF of 66% to 75% or >75% ( Table 1 ).
In the overall study group, no significant relation was evident between the EF and age, gender, body surface area, LV volumes, or maximum LV thickness or LV mass ( Tables 1 and 2 ). Also, no significant relation was evident between patients with HCM who had or had not received treatment with β blockers (72 ± 11% vs 70 ± 8%, respectively; p = 0.18), verapamil (73 ± 10% vs 71 ± 9%; p = 0.09), or disopyramide (73 ± 3% vs 71 ± 10%; p = 0.48). Patients with an EF <50% were more likely to have atrial fibrillation and to incur heart failure symptoms (New York Heart Association class II or III) than were the other 3 EF subgroups ( Table 1 ).
| Variable | Overall | LV EF (%) | p Value | |||
|---|---|---|---|---|---|---|
| <50% | 50–65% | 66–75% | >75% | |||
| Patients | 310 | 15 | 45 | 144 | 106 | |
| Age at study entry (y) | 42 ± 17 | 43 ± 17 | 40 ± 17 | 41 ± 18 | 46 ± 17 | 0.12 |
| Body surface area (m 2 ) | 1.9 ± 0.3 | 1.9 ± 0.3 | 2.0 ± 0.3 | 1.9 ± 0.3 | 1.9 ± 0.3 | 0.74 |
| Body mass index (kg/m 2 ) | 29 ± 7 | 29 ± 4 | 31 ± 9 | 28 ± 6 | 30 ± 7 | 0.10 |
| Left ventricular end-diastolic volume (ml) | 161 ± 45 | 197 ± 61 ⁎ † | 167 ± 48 | 162 ± 46 | 153 ± 36 | 0.003 |
| Left ventricular end-diastolic volume index (ml/m 2 ) | 83 ± 18 | 104 ± 29 ‡ | 84 ± 19 | 83 ± 17 | 79 ± 16 | <0.001 |
| Left ventricular end-systolic volume (ml) | 47 ± 24 | 114 ± 35 ‡ | 65 ± 19 ⁎ † | 47 ± 14 † | 31 ± 9 | <0.001 |
| Left ventricular end-systolic volume index (ml/m 2 ) | 24 ± 12 | 60 ± 19 ‡ | 32 ± 8 ⁎ † | 24 ± 6 † | 16 ± 4 | <0.001 |
| Left ventricular mass (g) | 203 ± 85 | 264 ± 142 ⁎ † | 214 ± 99 | 195 ± 78 | 200 ± 75 | 0.02 |
| Left ventricular mass index (g/m 2 ) | 103 ± 38 | 137 ± 64 ‡ | 107 ± 46 | 99 ± 34 | 102 ± 33 | 0.002 |
| Left ventricular mass/volume ratio | 1.3 ± 0.5 | 1.4 ± 0.6 | 1.3 ± 0.6 | 1.2 ± 0.4 | 1.3 ± 0.5 | 0.24 |
| Left ventricular ejection fraction (%) | 71 ± 10 | 42 ± 7 | 61 ± 4 | 71 ± 3 | 80 ± 3 | NA |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
