Background
Left ventricular noncompaction (LVNC) is characterized by a two-layered myocardium consisting of a noncompacted inner and a compacted outer layer. The ratio of the thicknesses of these two layers is a major diagnostic criterion, which is, however, often difficult to apply in clinical practice.
Methods
Transthoracic echocardiography was performed in 41 patients with LVNC, 41 patients with moderate or severe aortic valve stenosis (AVS), and 41 age-matched normal controls. The maximal systolic thicknesses of “noncompacta” and “compacta” were measured in standard short-axis views at the apical or midventricular level, in the segment with most prominent recesses (in patients with LVNC) or trabeculation (in patients with AVS and controls).
Results
The mean maximal systolic thickness of noncompacta was 1.8 ± 0.4 cm in patients with LVNC compared with 0.2 ± 0.1 cm in controls and 0.6 ± 0.02 cm in patients with AVS ( P < .0001). The mean maximal systolic thickness of compacta was lower in patients with LVNC (0.5 ± 0.1 cm) compared to controls (1.2 ± 0.2 cm; P < .0001) and patients with AVS (1.6 ± 0.06 cm; P < .0001). The maximal systolic thickness of compacta was ≤8.1 mm in patients with LVNC compared with >8.1 mm ( P < .0001) in controls and >8.1 mm in patients with AVS ( P < .0001). The ratio of the maximal systolic thickness of the indexed basal septum to that of the compacta was ≥0.64/m 2 in patients with LVNC compared to ≤0.62/m 2 in controls and ≤0.96/m 2 in patients with AVS.
Conclusions
Maximal systolic compacta thickness <8 mm is specific for LVNC and allows the differentiation of LVNC from normal hearts as well as those with myocardial thickening due to AVS. This observation may be particularly useful as an additional diagnostic criterion for preventing the overdiagnosis of LVNC.
Left ventricular noncompaction (LVNC) is caused by incomplete myocardial compaction during embryogenesis. LVNC occurs either as an isolated form or in association with other congenital heart diseases, and clinical manifestations are highly variable. Patients may be asymptomatic or present with symptoms of heart failure, ventricular arrhythmias, or thromboembolic events.
Morphologically, LVNC is characterized by prominent trabeculae and deep intertrabecular recesses communicating with the ventricular cavity, but not with the coronary circulation, which distinguishes them from myocardial sinusoids. Typically, the involved segments are located in the apical and inferolateral midventricular region of the left ventricle. These segments exhibit a two-layered structure resulting from the noncompacted inner layer and the compacted outer layer. Echocardiography is the gold standard for the diagnosis of LVNC and, because of its wide accessibility, remains the diagnostic modality of choice. Echocardiographic diagnostic criteria have been established by our group and others. These criteria include a thickened myocardium exhibiting a noncompacted inner layer and a compacted outer layer, a ratio of the systolic thickness of noncompacted to compacted layer > 2 in the parasternal short-axis view, color Doppler evidence of deep intertrabecular recesses filled with blood from the left ventricular cavity, and the typical distribution of affected segments in the midlateral, midinferior, and apical left ventricle. Overall, the criteria take into account the presence of prominent trabeculae and deep intertrabecular recesses leading to increased wall thickness and a two-layered appearance of the involved segments.
The current criteria, however, may be difficult to apply in both normal hearts and those with myocardial hypertrophy, because of technical obstacles such as oblique views. The latter may generate the false impression of deep recesses in any projection but present a particular problem in the parasternal short-axis view, as it is difficult to produce this view correctly in many individuals. Unfortunately, this view must be used for measuring the thickness of both myocardial layers to diagnose LVNC because of the orientation of the recesses.
Hence, additional echocardiographic criteria for distinguishing LVNC patients from both normal individuals and those with myocardial hypertrophy would be helpful. The aim of this study was to evaluate whether the thickness of the compacted layer could serve as a simple additional echocardiographic criterion to diagnose LVNC with higher accuracy in clinical practice.
Methods
Patients
Forty-one consecutive patients with definite diagnosis of LVNC were evaluated. The diagnosis of LVNC was established by echocardiography with strict adherence to previously published criteria comprising the following four conditions: (1) absence of coexisting cardiac abnormalities; (2) a two-layered myocardial wall with a compacted epicardial layer and a noncompacted endocardial layer, with a maximal end-systolic ratio of noncompacted to compacted layer > 2 in the parasternal short-axis view; (3) a predominant midlateral, midinferior, and apical location of affected segments; and (4) color Doppler evidence of deep intertrabecular recesses filled with blood from the left ventricular cavity. Myocardial wall thickness was measured at end-systole because the two layers are then easier to distinguish and because the ratio of the two layers was established by a side-by-side comparison of echocardiographic images with the corresponding pathologic specimens. Information regarding presenting symptoms and family history was obtained retrospectively from case notes and patient interviews. The control group comprised 41 age-matched and sex-matched subjects referred to echocardiography for the exclusion of a variety of abnormalities but who were found to have normal results. An additional group comprised 41 patients with concentric left ventricular hypertrophy due to moderate or severe aortic valve stenosis (AVS). Whereas both the LVNC and the control groups were examined in a retrospective manner, patients with AVS were analyzed prospectively to evaluate the novel diagnostic criteria in a patient group with left ventricular hypertrophy.
Echocardiography
Transthoracic two-dimensional (2D) echocardiography was performed using commercially available equipment (Acuson Sequoia C512, Siemens Medical Solutions USA, Inc., Malvern, PA; Philips iE33, Philips Medical Systems, Andover, MA; and GE Vivid 7, GE Healthcare, Milwaukee, WI). Images acquired in parasternal long-axis, parasternal short-axis, and apical views were analyzed offline on digitally stored loops. Standard measurements of left ventricular dimensions were obtained in accordance with the recommendations of the American Society of Echocardiography. In all patients and control subjects, echocardiographic measurements of myocardial layer thickness were made by two independent experienced observers, who determined wall thicknesses from digitally stored loops for this study. The duplicate measurements made by the two observers were averaged. The maximal systolic thicknesses of compacted and noncompacted layers were measured in standard parasternal short-axis views (2D) at the apical or midventricular level ( Figure 1 ). At the latter level, segments with papillary muscles were not used for measurements to avoid any confounding effects. In patients with LVNC, the maximal systolic thickness of the two layers was measured in the segment exhibiting the most prominent recesses. In controls and patients with AVS, the layers were analyzed in an analogous manner in the segment with the most prominent trabeculation. The thickness of the basal septum was measured in the standard parasternal long-axis view (M-mode and 2D), because this part of the myocardium is not affected by the disease process. The thickness of the basal septum was indexed to body surface area (BSA), because left ventricular wall thickness correlates with BSA; the thickness of the “compacta” was not indexed to BSA, because it is determined by the disease process. The ratio of indexed basal septum thickness to compacta thickness was calculated to analyze the influence of BSA and septal wall thickness on the proposed criterion.
A three-dimensional analysis of the left ventricle was performed in a healthy subject to illustrate that oblique slicing leads to overestimation of myocardial thickness. Three-dimensional (3D) full volumes of the left ventricle were recorded with four-beat acquisition using a Philips X5-1 transducer (Philips Medical Systems). Multiplanar reconstruction tools (QLAB 3DQ; Philips Medical Systems) were applied to measure left ventricular wall thickness in both a perpendicular view (0°) and an oblique view (20°) ( Figure 2 ).
To facilitate reading, both the noncompacted layer in patients with LVNC and the trabeculated layer in controls and patients with AVS are called “noncompacta” throughout the text, while both the compacted layer in patients with LVNC and the nontrabeculated layer in controls and patients with AVS are called “compacta” throughout the text.
Statistical Analysis
Normal distribution of values was tested using the Shapiro-Wilk test. Descriptive data for continuous variables are presented as mean ± SD. Continuous data were compared using analysis of variance followed by post hoc Tukey’s tests. All analyses were performed using SPSS for Windows version 19.0 (SPSS, Inc., Chicago, IL). A two-sided P value < .05 was established as the level of statistical significance for all tests.
Results
Patient Characteristics
A total of 41 patients (28 [68%] men) with LVNC, 41 age- and sex-matched controls, and 41 patients with hypertrophic myocardium due to moderate or severe AVS mean (left ventricular muscle mass index, 162 ± 7 g/m 2 ) were analyzed. The mean age at presentation was 36 ± 16 years in patients with LVNC and 36 ± 16 years in controls; the mean age in patients with AVS was slightly higher (48 ± 12 years; P < .001 vs patients with LVNC and controls). In patients with LVNC, the most commonly involved areas were the left ventricular apical, inferior midventricular, and lateral midventricular segments. The mean number of involved segments per patient was 2.8 ± 1.2 (range, 1–5). Thirty-eight patients (93%) had the isolated form of the disease, while three patients (7%) exhibited additional cardiac pathologies. Moderate aortic valve regurgitation was seen in one patient, while pulmonary valve disease was seen in two patients.
Thirty-eight patients (93%) with LVNC exhibited various degrees of left ventricular systolic dysfunction, with a mean left ventricular ejection fraction of 45% ( Figure 3 A). Patient characteristics are summarized in Table 1 .
| Clinical characteristic | Patients with LVNC ( n = 41) | Control ( n = 41) | Patients with AVS ( n = 41) | P |
|---|---|---|---|---|
| Age (y) | 36 ± 16 | 36 ± 16 | 48 ± 12 ∗ | ∗ <.005 |
| Body mass index (kg/m 2 ) | 22.7 ± 3.8 | 24.1 ± 4.6 | 27.5 ± 6.4 ∗ | ∗ <.009 |
| Men/women | 28/13 | 28/13 | 28/13 | NS |
| Embolic events/thrombus | 0 | 0 | 0 | NS |
| Left ventricular ejection fraction (%) | 45 (10–73) ∗ | 63 (55–74) | 58 (18–75) | ∗ <.001 |
| Pulmonary hypertension † | 4 | 0 | 0 | NS |
† Defined as tricuspid regurgitation maximal systolic flow velocity > 2.8 m/sec.
Effect of Oblique Views on Left Ventricular Wall Thickness
The effect of an oblique view on myocardial thickness of the left ventricle can be calculated by trigonometry when it is assumed that epicardium and endocardium are aligned in a parallel manner within the thin slice of myocardium containing the perpendicular and the oblique plane. Under these conditions, the length of line c crossing the myocardium obliquely at an angle α is equal to the length of line b crossing the myocardium perpendicularly divided by the cosine of the angle α ( c = b /cos α). If the length of line b equals 9.0 mm, the length of the line c will be equal to 9.6 mm if α = 20°, 11.7 mm if α = 40°, and 18.0 mm if α = 60°.
Left ventricular 3D full-volume acquisition permitted an evaluation of this theoretical calculation. Multiplanar reconstruction tools were applied to first determine the long axis of the left ventricle. A transverse plane, perpendicular to the long axis, was then introduced at the midventricular level ( Figure 2 ). Myocardial thickness was measured at the level of this transverse plane and equaled 9 mm both septally and laterally. The transverse plane was then tilted by 20°, and myocardial thickness was again determined. Under these conditions, its thickness equaled 10 mm both septally and laterally ( Figure 2 ).
Echocardiographic Parameters in Patients with LVNC
Left ventricular ejection fractions was significantly decreased in patients with LVNC (mean, 45%; range, 10%–73%) compared with controls (mean, 63%; range, 55%–74%; P < .0001) and patients with AVS (mean, 58%; range, 18%–75%; P < .0001) ( Figure 3 A). Left ventricular end-diastolic diameter (M-mode) was larger in patients with LVNC (5.7 ± 1.0 cm; P < .0001) and those with AVS (5.3 ± 0.7cm; P = .018) compared with controls (4.9 ± 0.4 cm) ( Figure 3 B), while left ventricular end-diastolic diameter did not differ significantly between patients with LVNC and those with AVS ( P = .093). Septal wall thickness (M-mode) was significantly increased in patients with AVS (1.4 ± 0.1 cm; P < .0001) compared with those with LVNC (0.9 ± 0.2 cm) and controls (0.9 ± 0.1 cm) ( Figure 3 C, Table 2 ). Similar observations were made for indexed septal wall thickness (0.7 cm/m 2 in patients with AVS vs 0.5 cm/m 2 in those with LVNC and 0.5 cm/m 2 in controls; P < .0001) ( Figure 3 D, Table 2 ). Additional 2D measurements of basal and midventricular septal thickness were performed in all patients to exclude any influence of the measuring technique ( Figure 4 ). There were no significant differences between 2D and M-mode measurements of the basal septal wall (M-mode: 0.8 ± 0.03 cm in controls, 0.9 ± 0.03 cm in patients with LVNC, and 1.3 ± 0.07 cm in patients with AVS; 2D: 0.9 ± 0.1 cm in controls, 0.9 ± 0.2 cm in patients with LVNC, and 1.4 ± 0.1 cm in patients with AVS; P = NS for all groups). When the septal wall was measured at the midventricular level (2D) in the parasternal long-axis view, there was no significant difference to the value obtained at the basal level (0.8 ± 0.08 cm in controls, 1.0 ± 0.03 cm in patients with LVNC, and 0.9 ± 0.03 cm in patients with AVS; P = NS for all groups).
| Parameter | Patients with LVNC | Controls | Patients with AVS | P |
|---|---|---|---|---|
| Total wall thickness (cm) | 2.3 ± 0.5 ∗ | 1.4 ± 0.2 | 2.2 ± 0.1 ∗ | ∗ <.0001 |
| Noncompacta thickness (cm) | 1.8 ± 0.4 ∗ | 0.2 ± 0.1 | 0.6 ± 0.02 | ∗ <.0001 |
| Compacta thickness (cm) | 0.5 ± 0.1 | 1.2 ± 0.2 | 1.6 ± 0.06 | ∗ <.0001 |
| Noncompacta thickness/compacta thickness | 3.5 ± 1.0 ∗ | 0.2 ± 0.1 | 0.4 ± 0.02 | ∗ <.0001 |
| Indexed septal wall thickness (M-mode) (cm/m 2 ) | 0.5 ± 0.2 | 0.5 ± 0.1 | 0.7 ± 0.03 | NS |
| Indexed septal wall thickness/compacta thickness (/m 2 ) | 1.0 ± 0.3 ∗ | 0.4 ± 0.2 | 0.5 ± 0.2 | <.0001 |
| Compacta thickness (mm) | ≤8.1 | >8.1 | >8.1 | <.0001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
