Diagnostic Value of Rigid Body Rotation in Noncompaction Cardiomyopathy


The diagnosis of noncompaction cardiomyopathy (NCCM) remains subject to controversy. Because NCCM is probably caused by an intrauterine arrest of the myocardial fiber compaction during embryogenesis, it may be anticipated that the myocardial fiber helices, normally causing left ventricular (LV) twist, will also not develop properly. The resultant LV rigid body rotation (RBR) may strengthen the diagnosis of NCCM. The purpose of the current study was to explore the diagnostic value of RBR in a large group of patients with prominent trabeculations.


The study comprised 15 patients with dilated cardiomyopathy, 52 healthy subjects, and 52 patients with prominent trabeculations, of whom a clinical expert in NCCM defined 34 as having NCCM. LV rotation patterns were determined by speckle-tracking echocardiography and defined as follows: pattern 1A, completely normal rotation (initial counterclockwise basal and clockwise apical rotation, followed by end-systolic clockwise basal and counterclockwise apical rotation); pattern 1B, partly normal rotation (normal end-systolic rotation but absence of initial rotation in the other direction); and pattern 2, RBR (rotation at the basal and apical level predominantly in the same direction).


The majority of normal subjects had LV rotation pattern 1A (98%), whereas the 18 subjects with hypertrabeculation not fulfilling diagnostic criteria for NCCM predominantly had pattern 1B (71%), and the 34 patients with NCCM predominantly had pattern 2 (88%). None of the patients with dilated cardiomyopathy showed RBR. Sensitivity and specificity of RBR for differentiating NCCM from “hypertrabeculation” were 88% and 78%, respectively.


RBR is an objective, quantitative, and reproducible functional criterion with good predictive value for the diagnosis of NCCM as determined by expert opinion.

Noncompaction cardiomyopathy (NCCM) is a myocardial disorder characterized by excessive and prominent trabeculations associated with deep recesses that communicate with the left ventricular (LV) cavity but not the coronary circulation. Although NCCM was included in the 2006 World Health Organization classification of primary cardiomyopathies, it remains subject to controversy because of a lack of consensus on its etiology, pathophysiology, diagnosis, and management. The normal left ventricle (LV) consists of obliquely oriented muscle fibers that vary from a smaller radius, right-handed helix at the subendocardium to a larger radius, left-handed helix at the subepicardium. The functional consequence of this three-dimensional helical structure is a cyclic systolic twisting deformation, resulting from opposite clockwise basal rotation and counterclockwise apical rotation. LV twist plays a pivotal role in the mechanical efficiency of the heart, making it possible that only 15% fiber shortening results in a 60% reduction in LV volume. Changes in LV twist have been reported in a variety of cardiac diseases. Our group recently reported nearly absent LV twist in a small group of patients with NCCM because of rotation of the basal and apical ends of the LV in the same direction, leading to rigid body rotation (RBR) with decreased circumferential-longitudinal shear deformation. We hypothesized that RBR may be a new objective functional diagnostic criterion for NCCM. The purpose of the present study was to further explore the diagnostic value of RBR in a larger group of patients with prominent trabeculations.


Study Participants

The study population consisted of 30 patients diagnosed before 2008 with NCCM by expert opinion (of whom 10 were included in a previous study on LV twist in NCCM) and 22 consecutive patients with prominent trabeculations (visual estimated end-systolic ratio of noncompacted to compacted layer >1.5) who underwent echocardiography in 2008, identified by one physician highly experienced with echocardiography (M.L.G.). All patients were in sinus rhythm and had good echocardiographic image quality that allowed for complete segmental assessment of LV rotation at both the basal and apical LV levels. During the enrollment of the 52 patients, 18 other patients (26%) were excluded because of suboptimal echocardiographic image quality not fulfilling this criterion. None of the patients had known coronary artery disease (excluded by coronary angiography), hypertension, or significant valvular heart disease. Patients were compared with 52 healthy age-matched and gender-matched control subjects without hypertension or diabetes and with normal left atrial dimensions, LV dimensions, and LV function. Furthermore, 15 patients with dilated cardiomyopathy (DCM) with LV volumes and ejection fractions comparable with those of the patients with NCCM were included as well. All subjects gave informed consent and the institutional review board approved the study.

Diagnostic Criteria for NCCM

Two methods were used to diagnose NCCM in the 52 patients. The first method was based on the echocardiographic diagnostic criteria for NCCM according to Jenni et al. : (1) a two-layered structure of the LV wall, with the end-systolic ratio of noncompacted to compacted layer >2 ( Figure 1 ); (2) finding this structure predominantly in the apical and midventricular areas; and (3) blood flow directly from the ventricular cavity into the deep intertrabecular recesses as assessed by Doppler echocardiography. The noncompacted-to-compacted ratio was quantitatively assessed, with the aid of electronic calipers, by one observer (M.L.G.) blinded to the results of LV twist and the expert opinion. In the second method, the diagnosis of NCCM was based on expert opinion. One clinical expert in NCCM diagnosis (K.C.) used, in addition to criteria of Jenni et al. , those of Finsterer and Stollberger, information on the history of the patient (including family history), and magnetic resonance imaging data but was also blinded to LV twist results. The 30 patients with previously established diagnoses of NCCM were revised according to these methods as well.

Figure 1

Example of echocardiographic short-axis images of both subjects with hypertrabeculation and patients with NCCM.


Two-dimensional grayscale harmonic images were obtained in the left lateral decubitus position using a commercially available ultrasound system (iE33; Philips Medical Systems, Best, The Netherlands), equipped with a broadband (1-MHz to 5-MHz) S5-1 transducer (frequency transmitted, 1.7 MHz; frequency received, 3.4 MHz). All echocardiographic measurements were averaged from three heartbeats. Measurements of LV dimensions, volumes, fractional shortening, and ejection fraction were obtained in accordance with the recommendations of the American Society of Echocardiography. The LV was divided into nine segments to describe the location of noncompacted segments: one apical, four midventricular, and four basal segments (with an anterior, inferoseptal, anterolateral, and inferior segment each).

To optimize speckle-tracking echocardiography, images were obtained at a frame rate of 60 to 80 frames/s. Parasternal short-axis images at the LV basal level (showing the tips of the mitral valve leaflets) with the cross-section as circular as possible were obtained from the standard parasternal position, defined as the long-axis position in which the LV and aorta were most in line with the mitral valve tips in the middle of the sector. To obtain a short-axis image at the LV apical level (just proximal to the level with end-systolic LV luminal obliteration), the transducer was positioned one or two intercostal spaces more caudal, as previously described by us. From each short-axis image, three consecutive end-expiratory cardiac cycles were acquired and transferred to a QLAB workstation (Philips Medical Systems) for offline analysis.

Speckle-Tracking Analysis

Analysis of the data sets was performed using QLAB version 6.0, which was recently validated against magnetic resonance imaging for assessment of LV twist. To assess LV rotation, six tracking points were placed manually (after gain correction) on the midmyocardium on an end-diastolic frame in each parasternal short-axis image. In areas of hypertrabeculation, the tracking points were placed in the inner to midsection of the compacted part of the muscle. Tracking points were separated about 60° from one another and placed at 1 o’clock (30°, anteroseptal insertion into the LV of the right ventricle), 3 o’clock (90°), 5 o’clock (150°), 7 o’clock (210°), 9 o’clock (270°, inferoseptal insertion into the LV of the right ventricle), and 11 o’clock (330°) to fit the total LV circumference. After positioning the tracking points, the program tracked these points on a frame-by-frame basis using a least squares global affine transformation. The rotational component of this affine transformation was then used to generate rotational profiles.

Data were exported to a spreadsheet program (Excel; Microsoft Corporation, Redmond, WA) to determine LV peak systolic rotation during the isovolumic contraction phase (Rot early ), LV peak systolic rotation during ejection (Rot max ), and instantaneous LV peak systolic twist (Twist max , defined as the maximal value of instantaneous apical systolic rotation − basal systolic rotation). Counterclockwise rotation and twist as viewed from the apex were expressed as a positive value, and clockwise rotation and twist were expressed as a negative value. End-systole was defined as the point of aortic valve closure. In the present study, different LV rotation patterns were recognized ( Figure 2 ):

  • 1.

    Normal rotation

    • A.

      Completely normal rotation, characterized by initial counterclockwise and end-systolic clockwise basal rotation and initial clockwise and end-systolic counterclockwise apical rotation

    • B.

      Partly normal rotation, characterized by end-systolic clockwise basal rotation and end-systolic counterclockwise apical rotation but absence of either or both initial counterclockwise basal rotation or initial clockwise apical rotation

  • 2.


    • A.

      Clockwise RBR, characterized by clockwise basal and apical rotation throughout systole

    • B.

      Counterclockwise RBR, characterized by counterclockwise basal and apical rotation throughout systole

    • C.

      Initial clockwise followed by counterclockwise RBR

    • D.

      Initial counterclockwise followed by clockwise RBR

Figure 2

Schematic graphs of LV rotation patterns in healthy controls, subjects with hypertrabeculation, and patients with NCCM.

Statistical Analysis

Data are expressed as mean ± SD. Variables were compared using Student’s t tests or χ 2 tests as appropriate. P values < .05 were considered statistically significant. Intraobserver and interobserver variability for LV twist were 5 ± 4% and 9 ± 4%, respectively, in line with our previous study of the feasibility of LV twist measurement using speckle-tracking echocardiography. To test the reproducibility of LV rotation patterns in the current study, speckle-tracking analysis was repeated by a different physician (F.K.) in all patients. There was 100% agreement on the observed LV rotation patterns.


Characteristics of the Study Population

Revision of the 30 patients with previously established diagnoses of NCCM led to confirmation of the diagnosis in 29 by the criteria of Jenni et al. and in all 30 by expert opinion. Of the 22 patients with various degrees of hypertrabeculation, seven were classified as having NCCM by the criteria of Jenni et al. and four by expert opinion. The remaining patients were classified as “subjects with hypertrabeculation.” So, in total, 36 patients were diagnosed as having NCCM by the criteria of Jenni et al. and 34 by expert opinion. In four patients, the expert diagnosis was discrepant from the criteria of Jenni et al. , on the basis of information about race, family history, LV function, and the results of magnetic resonance imaging. Clinical and conventional echocardiographic characteristics of the study population are shown in Table 1 .

Table 1

Characteristics of the study population

Patients with NCCM Subjects with hypertrabeculation
Jenni et al. criteria Expert opinion Jenni et al. criteria Expert opinion DCM Controls
Variable ( n = 36) ( n = 34) ( n = 16) ( n = 18) ( n = 15) ( n = 52)
Clinical data
Age (y) 43 ± 15 44 ± 14 48 ± 18 46 ± 19 41 ± 14 44 ± 15
Men 19 (53%) 18 (55%) 8 (50%) 9 (47%) 7 (47%) 27 (52%)
QRS duration (ms) 105 ± 23 106 ± 27 111 ± 26 108 ± 25 115 ± 30 88 ± 8
Bundle branch block (left/right/nonspecific) 5/0/2 5/0/2 3/0/2 3/0/2 4/0/1 0/0/0
Echocardiographic data
LVEDD (mm) 57 ± 8 57 ± 7 53 ± 6 54 ± 7 63 ± 4 50 ± 6
LVESD (mm) 45 ± 8 45 ± 9 40 ± 10 41 ± 10 50 ± 8 34 ± 6
LV fractional shortening (%) 22 ± 7 22 ± 8 24 ± 8 24 ± 9 21 ± 6 32 ± 7
LVEDV (mL) 149 ± 50 150 ± 53 146 ± 53 145 ± 48 160 ± 55 115 ± 23
LVESV (mL) 90 ± 42 89 ± 41 81 ± 45 82 ± 46 110 ± 45 44 ± 15
LVEF (%) 42 ± 14 40 ± 12 44 ± 17 45 ± 18 38 ± 9 62 ± 7
Ratio of noncompacted to compacted layer 2.6 ± 0.5 2.6 ± 0.5 1.7 ± 0.3 1.7 ± 0.3 NA NA
Number of noncompacted segments 3.9 ± 2.5 3.8 ± 2.4 2.9 ± 1.6 3.0 ± 1.8 0 ± 0 0 ± 0
Absolute Twist max (°) 3.9 ± 2.2 § 4.1 ± 2.2 § 7.1 ± 4.9 6.9 ± 5.4 6.1 ± 2.3 10.1 ± 2.3

Data are expressed as mean ± SD or as number (percentage).

LVEDD , LV end-diastolic dimension; LVEDV , LV end-diastolic volume; LVEF , LV ejection fraction; LVESD , LV end-systolic dimension; LVESV , LV end-systolic volume; NA , not available.

P < .05, P < .01, and P < .001 versus patients with NCCM and subjects with hypertrabeculation; § P < .05 versus subjects with hypertrabeculation.

LV Rotation and Twist in Normal Subjects and Patients With DCM

In all but one normal subject, initial counterclockwise rotation at the LV basal level and initial clockwise rotation at the LV apical level could be identified (basal Rot early 2.0 ± 1.2°, and apical Rot early −0.8 ± 0.6°, respectively). Furthermore, peak end-systolic rotation was always in a clockwise direction at the LV basal level and in a counterclockwise direction at the LV apical level (basal Rot max −3.6 ± 1.8°, and apical Rot max 7.2 ± 2.9°, respectively), leading to a Twist max of 10.1 ± 2.3°. Although initial counterclockwise rotation at the LV basal level and initial clockwise rotation at the LV apical level were absent (LV rotation pattern 1B) in nine patients with DCM (60%) none of the patients with DCM showed RBR ( Table 2 ).

Table 2

LV rotation patterns in patients with NCCM, subjects with hypertrabeculation, patients with DCM, and controls

Patients with NCCM Subjects with hypertrabeculation
Jenni et al. criteria Expert opinion Jenni et al. criteria Expert opinion Patients with DCM Controls
Rotation pattern ( n = 36) ( n = 34) ( n = 16) ( n = 18) ( n = 15) ( n = 52)
1A 0 0 4 4 6 51
1B 6 4 8 10 9 1
Total 6 4 12 14 15 52
2A 13 13 1 1 0 0
2B 2 2 0 0 0 0
2C 1 1 1 1 0 0
2D 14 14 2 2 0 0
Total 30 30 4 4 0 0

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jun 15, 2018 | Posted by in CARDIOLOGY | Comments Off on Diagnostic Value of Rigid Body Rotation in Noncompaction Cardiomyopathy

Full access? Get Clinical Tree

Get Clinical Tree app for offline access