LV Non-compaction

Table of Contents

Part 1: Dilated Cardiomyopathy

Part 2: Hypertrophic Cardiomyopathy

Part 3: Restrictive Cardiomyopathy

Part 4: Arrhythmogenic Cardiomyopathy

Part 5: LV Non-compaction

Part 6: The Athlete's Heart

Introduction

Left ventricular non-compaction (LVNC) is a rare disorder that is considered an ‘unclassified cardiomyopathy’ by the European Society of Cardiology. Several different gene mutations related to LVNC have been identified, involving sarcomeric, cytoskeletal, Z-line, ion channel, mitochondrial, and signalling proteins. However, there is broad genetic overlap between LVNC and other inherited cardiac diseases such as dilated cardiomyopathy and hypertrophic cardiomyopathy. LVNC could also be part of multisystemic genetic entities such as Barth syndrome, or accompany congenital heart defects. LVNC is a morphological description of a spongy appearing myocardium, characterised by a normal or thin outer layer of compacted myocardium and an inner non-compacted layer with prominent trabeculae and intratrabecular recesses that communicate with the left ventricle (LV). It is most frequent in the apex of the LV but may be seen in both ventricles or be rarely confined to the right ventricle. The standard electrocardiogram does not contribute to a specific diagnosis of LVNC although it is generally abnormal in conditions associated with LVNC. Imaging modalities including echocardiography and cardiac MRI have contributed significantly to the identification of LVNC. The diagnosis of LVNC relies on demonstration of hypertrabeculation of the endocardial part of the myocardium, a normal or thin compacted myocardial layer, and the presence of deep intratrabecular recesses that communicate with the LV cavity. Although there is no consensus on the specific definition of LVNC, most imaging criteria depend upon measurement of the ratio between non-compacted (NC) and compacted myocardium (C). Studies using cardiovascular magnetic resonance imaging have considered a ratio of NC/C above 2.3 measured in end-diastolic long-axis view as diagnostic of LVNC, while echocardiography studies have used a ratio above two measured in end-systolic short-axis view to define the abnormality. The presentations on echocardiography (Erwan Donal) and MRI (Ana Almeida and Steffen Peterson) discuss this in extent.

See references at the bottom of the page

Presentations

• Echocardiography
• Cardiovascular Magnetic Resonance
• Multi-modality Imaging

Resources

References (introduction)

ESC CardioMed (3 edn)
Edited by A. John Camm, Thomas F. Lüscher, Gerald Maurer, and Patrick W. Serruys
Chapter: Left ventricular non-compaction: genetics and embryology
Pablo Garcia-Pavia and Fernando Dominguez
DOI:10.1093/med/9780198784906.003.0362_update_001
Chapter: Left ventricular non-compaction: diagnosis and clinical management.
Jens Mogesen and Torsten B Rasmussen
DOI:10.1093/med/9780198784906.003.0364_update_001

Presentations from recent ESC/EACVI congresses

Specific access rules may apply

EACVI webinars

Specific access rules may apply

• LVNC: the boundary of cardiovascular imaging

EACVI webinar
Thursday 14 May 2020 from 18:00 to 19:00 CESTProf. Cristina Basso
University of Padua Medical School
Padua, Italy

Dr. Juan Pablo Kaski
Great Ormond Street Hospital and University College London
London, UK

Host
Prof. Giovanni Di Salvo
University of Padua Medical School
Padua, Italy

Guidelines / position statements

• The EACVI Textbook of Echocardiography, 2nd Edition, OUP
  Chapter 45 Isolated ventricular non-compaction
  Frank Weidemann, Gilbert Habib and Clara Vazquez
• The ESC Textbook of Cardiovascular Medicine, 3th edition, OUP
  Chapter 32.16 Left ventricular non-compaction: genetics and embryology
  Pablo Garcia-Pavia and Fernando Dominguez
  Chapter 32.17 Left ventricular non-compaction: diagnosis and clinical management
  Jens Mogensen and Torsten B Rasmussen
• Classification of the cardiomyopathies: a position statement from the European Society of Cardiology Working Group on Myocardial and Pericardial Diseases
  Perry Elliott, Bert Andersson, Eloisa Arbustini, Zofia Bilinska, Franco Cecchi, Philippe Charron, Olivier Dubourg, Uwe Kühl, Bernhard Maisch, William J. McKenna, Lorenzo Monserrat, Sabine Pankuweit, Claudio Rapezzi, Petar Seferovic, Luigi Tavazzi, Andre Keren
  European Heart Journal, Volume 29, Issue 2, January 2008, Pages 270–276

Litterature review

Left Ventricular Noncompaction & Hypertrabeculation in CMR Research – Literature Review

Dr Simon Woodbridge

Left ventricular non-compaction is a recognised cardiomyopathy within the cardiovascular community characterised by a high level of left ventricular trabeculation (with deep intertrabecular recesses) teamed with a thin compacted epicardial layer. It remains to be classified as a genetic cardiomyopathy by the AHA, and ‘unclassified’ by WHO. With regards to its clinical significance today, an article by Arubustini et al. in JACC in 2016 (1) summarises this well. They outline the prevailing debate regarding how LVNC may be diagnosed, and whether its physiology influences LV function intrinsically enough to be considered a cardiomyopathy. The report concluded (using current study data at the time) that LVNC is more likely to present as a characteristic of a ‘wider’ cardiomyopathy, rather than exhibiting a direct causal link with the mechanical or electrical dysfunction that a cardiomyopathy would cause (as per the ESC definition of a cardiomyopathy).

The debate continues, and the recent meta-analysis by Aung et al. (2) has concluded that LVNC exbihits similar risks of cardiovascular & all-cause mortality, thromboembolic complications, and ventricular arrhythmias as patients with dilated cardiomyopathy. This paper also reaffirmed the level of heterogeneity between existing studies investigating the clinical outcomes of LVNC. I will concentrate more on what developments have been made with regards to measurement of trabeculation within LVNC, and how this might impact how it has been diagnosed.

Measurement of trabeculation

Petersen Method - 2005:

The non-compaction to compaction ratio (NC/C) is a measurement method coined by Petersen et al. in 2005 (3). It involves a linear measurement of the thickness of the non-compacted (trabeculated) and compacted myocardium in long-axis views of the left ventricle at the same location. The measurement is taken after qualitative assessment for the presence of trabeculation, defined in this context as the visual identification of two layers of the myocardium, represented by a difference in signal intensity between the two layers (the inner layer being lighter, and hence less compacted). Typically, the segment with the most pronounced trabeculation upon visual inspection is then measured for each of the basal, mid and apical regions, if present. The true apex is excluded from analysis, as the compacted myocardium is usually thinner at this point, leading to inaccurate ratios.
This method is very quick, and easy to teach to others – this makes it beneficial for data collection from large cohorts to produce highly powered results. However, reproducibility is potentially an issue, due to the segments chosen for measurement being at the discretion of the observer, and less ‘obvious’ scans may produce more inter-observer differences. Also, the ratio can be confounded by the thickness of the compacted myocardium, for instance in cases of left ventricular hypertrophy perhaps due to chronic hypertension. For example, non-compacted trabecular thickness may be high with a value of 9mm, however, due to hypertension-related hypertrophy, the compacted thickness may also be 9mm. The heart would therefore be highly trabeculated, but with a low NC/C ratio of one, misrepresenting the trabeculation. The presence of LVNC was described as an NC/C ratio of greater than 2.3.

Stacey Method - 2013:

This method, described by Stacey et al. in 2013 (4) is similar to the Petersen method in nature, however this acquires the measurement in end-systole, using short axis images. LVNC is defined here as an NC/C ratio greater than two. Stacey’s paper concluded that measuring NC/C ratio to diagnose LVNC in end-systolic frames had a stronger relationship with LV function and identification of heart failure than the Petersen method. Again, the same limitations exist, with the potential variability in qualitative location identification for ratio measurement presenting in a different plane (short vs. long axes).

Jacquier Method - 2010:

The percentage trabeculated mass method was introduced by Jacquier et al. in 2010 (5), and analyses trabeculation in the left ventricle based on short-axis CMR images, taken from base to apex at end diastole. It uses semi-automatic contouring to determine the global myocardial mass (being the compacted myocardium, papillary muscles and trabeculation observed) as well as the compacted myocardial mass (being the compacted myocardium and the papillary muscles) across all short-axis slices, and uses the below formula to calculate trabeculation represented as a proportion of the global myocardial mass:

Percentage trabeculated mass (%) = ((Global myocardial mass – Compacted myocardial mass)/Global myocardial mass) x 100

This method evaluates the trabecular extent globally across the left ventricle, providing a more comprehensive measurement than the NC/C ratio. This global measurement, when compared to the manual choice of the long-axis trabeculated regions to measure with the NC/C ratio, should, in theory, produce results with higher reproducibility between observers. However, based on the anecdotal reports from analysts with experience in Jacquier’s method, the inter- and intra-observer variability can be high due to difficulties in delineation of trabeculated region using the currently-available contouring tools. In addition, by its nature, the method is more time consuming than the NC/C ratio, which can be a hindrance when collecting data from a large cohort. Other limitations include possible misidentification of papillary muscle as trabeculation (if the distinction is unclear, papillary muscle is treated as trabeculation), leading to overestimations of percentage trabeculation values. Also, in high levels of trabeculation, measuring the mass does not take into account the exclusion of possible blood pools formed by the trabeculated region, again providing overestimations of trabeculation. In Jacquier’s study, a percentage of over 20% was considered as a diagnostic cut-off for diagnosing LVNC.

Captur Method - 2013:

Trabecular structure follows fractal biology – ‘fractals’ are mathematical sets which exhibit a complex self-repeating pattern regardless of scale, present in many biological structures. Captur et al. (6) derived a method of measuring trabeculation extent based on this fractal biology. Like the percentage trabeculated mass, the measurements are taken from end-diastolic short-axis views, excluding the true apex. Fractal analysis is not a measure of thickness or mass, but rather the complexity of the trabeculation observed.

A unitless measure index was created (fractal dimension, FD), which ranges from one to two. As trabeculation is more complex than a simple straight line (one dimension, 1D), the value must be over one, but as trabeculation does not completely fill the two-dimensional (2D) space provided by a short-axis slice of the left ventricle, the value must be less than two. The FD measurement therefore evaluates the extent to which trabeculation fills a 2D space, consistently represented as a value anywhere between one and two. Fractal analysis builds upon the detail provided by the NC/C ratio and the percentage trabeculated mass. It uses a method of ‘box counting’ – after excluding the compacted myocardium, a box grid of known scale (spacing between the boxes) is laid over the short-axis slice, and the number of boxes containing trabeculation, or any matter, is counted. As the scale increases and the box spacing becomes larger, the number of boxes containing matter will exponentially decrease at a rate which is equal to the FD value, after logarithmic quantification. The method does not take the compacted wall thickness into account as a contribution to trabecular extent.

While the omission of compacted wall thickness from the analysis means there is no confounding factors such as those affecting the NC/C ratio, it however loses the usefulness of this measurement in the context of LVNC diagnosis, where the compacted wall is expected to be thinner.

Bricq method - 2015:

Bricq et al. in 2015 (7) published a study which utilised automatic contouring of noncompacted, endocardial and epicardial boundaries, in similar nature to Jacquier’s method, but a further semiautomatic step in analysis allowed for the removal of blood pools, which was not factored into Jacquier’s method. This was performed using thresholding programmed to distinguish between bright intensity blood and the darker intensity trabecular compartment. This same thresholding was also used to distinguish the papillary muscle, and this was propagated across all slices of the LV. This study is of particular note in that it directly compared contours and trabeculation between CMR slices and histological sections of mice with hypertrabeculated hearts (used as a validation cohort for the method). The comparison demonstrated good correlation of mass values between CMR radiology and histology.

General comments

In the context of CMR and excluding advances in automation, there seems to be little evidence of new developments in trabeculation measurement modality since around 2015. CMR introduced superior spatial resolution over echocardiography (also considered a diagnostic tool for LVNC, with the classic trio of the Chin, Jenni and Stöllberger criteria used to quantify trabeculation) but the gold standard method of measurement is still very much up for debate.

There remains little evidence as to whether hypertrabeculation in and of itself has any prognostic or pathophysiological effect. Indeed, some have argued that LVNC as a cardiomyopathy should be abandoned, and rather deemed ‘excessive trabeculation’ (8)(9). This could be more appropriately grouped with other cardiomyopathies as a feature for which there appears to be stronger evidence that it may carry prognostic weight. 

I would argue that this isn’t necessarily the end of the story, as the general impression of prognostic investigations into LVNC is that they are very heterogeneous (as mentioned previously) and there is yet to be a high powered prospective cohort study (as mentioned by Petersen & Neubauer in 2017 (10)) to investigate its prognostic value further. But before we get to this point, the time seems apt for researchers to decide on a single gold standard of LV trabeculation measurement. This will allow us to concentrate first on ‘excessive trabeculation’, what this actually represents clinically, and then introduce non-compaction as a secondary investigation (this should be relatively straightforward, with accurate measurement of the compacted endocardium easier to accomplish). The Bricq method of measurement appears to show the most comprehensive measurement of LV trabeculation so far – it tangibly measures trabeculation extent with efforts to remove ‘noise’ of blood pools, it measures across the entire LV (with scope for regional submeasurements) and hence trabeculation can be interpreted as a three-dimensional value.

Conclusion

Put simply, given that LVNC as a clinical diagnosis remains unclear, we should at least enable the method of trabeculation measurement to be robust, and a consensus reached on this, to facilitate further necessary research into LVNC. This would be best met by high automaticity (for rapid analysis), accuracy, and reproducibility. Developments in automatic qualitative analysis (such as that of image intensity to distinguish levels of compaction) and machine learning to continually improve these measurements should now enable such a large scale prospective cohort study using one of the suite of measurement methods described in this review.

References

(1) Arbustini E, Favalli V, Narula N, Serio A, Grasso M. Left Ventricular Noncompaction: A Distinct Genetic Cardiomyopathy? J Am Coll Cardiol. 2016 Aug 30;68(9):949–66.

(2) Prognostic Significance of Left Ventricular Noncompaction | Circulation: Cardiovascular Imaging [Internet]. [cited 2020 Jul 20]. Available from: https://www.ahajournals.org/doi/10.1161/CIRCIMAGING.119.009712

(3) Petersen SE, Selvanayagam JB, Wiesmann F, Robson MD, Francis JM, Anderson RH, et al. Left ventricular non-compaction: insights from cardiovascular magnetic resonance imaging. J Am Coll Cardiol. 2005 Jul 5;46(1):101–5.

(4) Stacey RB, Andersen MM, St Clair M, Hundley WG, Thohan V. Comparison of systolic and diastolic criteria for isolated LV noncompaction in CMR. JACC Cardiovasc Imaging. 2013 Sep;6(9):931–40.

(5) Jacquier A, Thuny F, Jop B, Giorgi R, Cohen F, Gaubert J-Y, et al. Measurement of trabeculated left ventricular mass using cardiac magnetic resonance imaging in the diagnosis of left ventricular non-compaction. Eur Heart J. 2010 May;31(9):1098–104.

(6) Captur G, Muthurangu V, Cook C, Flett AS, Wilson R, Barison A, et al. Quantification of left ventricular trabeculae using fractal analysis. Journal of Cardiovascular Magnetic Resonance. 2013 May 10;15(1):36.

(7) Bricq S, Frandon J, Bernard M, Guye M, Finas M, Marcadet L, et al. Semiautomatic detection of myocardial contours in order to investigate normal values of the left ventricular trabeculated mass using MRI. Journal of Magnetic Resonance Imaging. 2016;43(6):1398–406.

(8) Anderson RH, Jensen B, Mohun TJ, et al. Key Questions Relating to Left Ventricular Noncompaction Cardiomyopathy: Is the Emperor Still Wearing Any Clothes?. Can J Cardiol. 2017;33(6):747-757. doi:10.1016/j.cjca.2017.01.017

(9) Aung N, Zemrak F, Petersen SE. Left Ventricular Noncompaction, or Is It? J Am Coll Cardiol. 2016 Nov 15;68(20):2182–4.

(10) Petersen Steffen E., Neubauer Stefan. Excessive Trabeculations and Prognosis. Circulation: Cardiovascular Imaging. 2017 Sep 1;10(9):e006908.