Author + information
- Bengt Johansson, MD, PhD,
- Alicia M. Maceira, MD,
- Sonya V. Babu-Narayan, BSc, MRCP,
- James C. Moon, MRCP, MD,
- Dudley J. Pennell, MD, FRCP, FESC, FACC and
- Philip J. Kilner, MD, PhD⁎ ()
- ↵⁎Cardiovascular Magnetic Resonance Unit, Royal Brompton Hospital, Sydney Street, SW3 6NP, London, United Kingdom
To the Editor:New imaging techniques allow the recognition of structural features that have previously been overlooked. Cardiovascular magnetic resonance (CMR) steady-state free-precession (SSFP) cine imaging gives good contrast between the blood and the myocardium and good in-plane resolution of thin structures or interstices as long as their boundaries extend parallel to the elongated voxels, orthogonal to the plane of acquisition. We have noticed appearances indicative of discrete clefts or fissures in otherwise compact myocardium of the left ventricle (LV) in long-axis cine acquisitions in healthy volunteers as well as patients. Germans et al. (1) recently reported structural abnormalities, which they termed crypts, in the basal and midinferoseptal walls of 13 out of 16 carriers of hypertrophic cardiomyopathy mutations, but none in the 16 healthy volunteers that they studied.
We analyzed the previously acquired CMR studies in 399 individuals in 5 groups who had been studied for reasons other than the visualization of ventricular clefts. They included 120 asymptomatic healthy volunteers (n = 120; 10 men and 10 women in each of the 6 age deciles from 20 years to 80 years) whose LV measurements have been previously reported (2) and 4 patient groups: 91 with hypertrophic cardiomyopathy (HCM), 44 with systemic hypertension, 104 with repaired tetralogy of Fallot (rToF), and 40 with surgically relieved congenital pulmonary valve stenosis (rPS). We included 1 additional patient known to have a small muscular ventricular septal defect and suspected HCM, identified as having clefts at the time of his CMR study, to illustrate their location and orientation in images aligned to transect them in 3 orthogonal orientations. The volunteers and patients had given informed consent, and retrospective analysis and use of data were approved by the local ethics committee.
Using a 1.5-T whole-body CMR scanner (Siemens Sonata, Erlangen, Germany), acquisitions included SSFP cines, voxel dimensions typically 1.3 × 2 × 7 mm, in 3 LV long-axis planes—a 2-chamber vertical long axis (VLA) plane located orthogonal to transaxial scouts through the mitral valve and the LV apex, a 3-chamber located through the apex and the mitral and aortic valves, and a 4-chamber located through the mitral and tricuspid valves and the apex—and a short-axis stack for measurements of ventricular volume and mass (2). We defined clefts as discrete approximately V-shaped extensions of blood signal penetrating >50% of the thickness of adjoining compact myocardium in long-axis views, the cleft tending to narrow or occlude in systole, without local hypokinesia or dyskinesia.
Basal inferior myocardial clefts were visible in the basal inferior wall of the LV as seen in the VLA cine images in 7 out of 120 volunteers (6%), 5 male and 2 female. All 7 are illustrated in Figures 1Ato 1G. They appeared to be single in 6 and paired in 1 (Fig. 1F). Clefts were found in 5 out of 91 HCM patients (5.5%), 5 out of 44 hypertensives (11.4%), 1 out of 104 rToF patients (1%), and 9 out of 40 rPS patients (22.5%). They were usually single, with small additional clefts faintly visible in a minority of cases. No evidence of relationships between cleft visibility and age, gender, or LV function or mass was found in the volunteer or patient groups.
Septal clefts were seen in the mid to apical parts of the interventricular septum in the 3- or 4-chamber cines in 24 (6%) of the 399 individuals studied, in 6 of the 120 volunteers, 13 of the 104 rToF patients, and 5 of the 40 rPS patients. Most were adjacent to the insertions of trabecular bands into the right ventricular side of the septum, as in Figure 1L.
Acquisitions aligned specifically with respect to inferoseptal and anteroseptal clefts are shown in Figure 2.
In conclusion, single or paired clefts were visible in the basal inferior wall of the LV and the interventricular septum in some volunteers as well as in patients, even when only the routine long-axis cines were reviewed, and these planes may not have shown all of the clefts present. The VLA slice typically intersects the basal inferior wall slightly to the left of the inferoseptal junction, the region in which crypts were identified by Germans et al. (1). Comparison of our illustrations with theirs suggests that we may have described related structural features. The region of insertion between the free walls of the left and right ventricles and the interventricular septum is known to be subject to myocardial disarray, described postmortem in the hearts of some healthy individuals as well as those with HCM, with adjacent deep tissue clefts having been identified in the latter (3). Disarray may be a histologic manifestation of the crossing and interdigitation of myocytes, which is likely to be structurally and mechanically advantageous in these specific regions. But the interdigitation of elongated quasicylindrical structures is not compatible with perfect packing without either local compression or interstices adjacent to the points of crossing. Whatever the underlying cause, however, cine CMR appearances leave little doubt that intramyocardial fissures or clefts occur in certain typical locations in healthy volunteers. Awareness of this may avert unwarranted further investigation and anxiety.
For accompanying videos, please see the online version of this article.
- American College of Cardiology Foundation