An assessment of early cardiomyopathy in Autosomal Dominant Emery-Dreifuss Muscular Dystrophy

Gillian C Smith1, Sanjay K Prasad1, Maria Kinali3, Sharmeen Masood4, Guan-Zhong Yang4, Francesco Muntoni3, Petros Nihoyannopoulos2, Dudley J Pennell1

CMR Unit, Royal Brompton Hospital, London, UK1, Cardiology Dept., Hammersmith Hospital London UK2, Dubowitz Neuromuscular Centre, Hammersmith Hospital London UK3, Royal Society/Wolfson Medical Image Computing laboratory, Imperial College4

Introduction: Emery-Dreifuss (EDMD) muscular dystrophy comprises a phenotypically and genotypically heterogeneous familial disorder, initially characterised as an X-linked disease caused by mutations in the gene encoding the nuclear protein emerin. Recently the gene for EDMD2 has been identified which encodes lamins A and C. Skeletal muscle symptoms are often mild but cardiac involvement may be profound, usually manifesting by the second or third decade. These abnormalities range from ventricular arrhythmias and conduction defects to cardiomyopathy. The early detection of cardiac abnormalities is important therefore to select the best therapeutic strategy. Cardiovascular magnetic resonance (CMR) has been used in patients with congenital myotonicdystrophy and reported to identify changes that might suggest the early presence of fibrosis. Our objective was to determine whether atrial or left ventricular dysfunction could be detected preclinically, and whether fibrotic myocardial changes accompany or precede the conduction system damage in EDMD2.

Methods: Eight consecutive genetically proven EDMD2 patients were enrolled in the study. As the disease incidence is low and many patients have implanted pacemakers the potential study population was correspondingly low. Six patients were male, the age range from 7-42 years (mean 18.5). An age matched control group was obtained from eight unaffected volunteers. The CMR examination was performed using a 1.5 T scanner (Magnetom Sonata, Siemens AG Medical Engineering, Erlangen, Germany) using front and back surface coils and prospective ECG triggering during breath-hold. Functional imaging was performed using a steady state free precession (SSFP) sequence. From serialshort axis projections (Figure 1)

accurate volume and mass measurement used semi-automated segmentation (CMRtools, Cardiovascular Imaging Solutions, London, UK). Turbo spin-echo imaging with T1 and T2 weighting with and without fat suppression was used to show the presence of fibro-fatty replacement. Gadolinium-DTPA was given as a peripheral intravenous bolus (0.1 mmol./kg bodyweight) to test whether fibrosis plays a role in cardiac involvement in EDMD. Contrast enhanced images were acquired using a segmented inversion recovery sequence. Using the harmonic phase tagging method (HARP) we acquired 2 long-axis and 3 short-axis projections to give good coverage of all myocardial segments to depict longitudinal, radial and circumferential motion. We calculated minimal principal strain for controls and study patients (defined as maximal shortening parallel to the fibre direction reflecting circumferential strain patterns)In addition to the full 2-DE/Doppler examination we performed spectral and colour coded tissue imaging. The latter showing spatial and temporal velocity differences during systole and early and late diastole from which myocardial velocity gradients could be calculated

Results: Table 1 outlines patient clinical and genetic profiles.They were all asymptomatic from the cardiac viewpoint and none was on any medication. Table 2 shows all CMR derived measurements. Left and right ventricular ejection fractions were within normal limits for both patient and control groups. There was no significant difference between the corrected atrial and ventricular volumes and LV mass between the two groups. Echo result are described in Table 3. Patients had normal left ventricular size and function and there was no significant difference between patients and controls. Mitral E and A wave diastolic velocities were similar for both groups as were the annular E to A wave ratios Pulse tissue Doppler velocities were normal in all mitral annular regions and were similar between patients and controls. The mitral E to annular E’ wave ratios was also normal Systolic myocardial velocity gradients were similar between patients and controls. The early diastolic gradient was lower however in patients than in controls suggesting early myocardial dysfunction (4 ± 1.2 s-1 vs. 7.1 ± 2.7 s-1,p=0.02)

Mean left ventricular mass when corrected for body surface area was below the published normal value of 87±12 g/m2. The corrected mass for the control group was also below the normal range. Some of these differences may be explained by the relative youth of both patient and control samples when compared with the age range of the previously described group (8-55 yrs, mean 28±9) versus 7-42 yrs mean 18.5 and 6-30 yrs mean 15. No patient showed areas of late gadolinium enhancement or signs of fibro-fatty replacement. There was however a significant reduction in inferior wall strain as determined by HARP tagging at -0.062 + 0.024 versus -0.094 + 0.032 in the control group (p=0.048)

NormalNormalNormalNormalNormalNormalNormalNormal2-D echo

AV blockNormalNormalNormal1st degree AV block

NormalNormalNormalHolter

Atrial flutterNormalNormalNormal1st degree AV block

NormalNormalNormalECG

ModerateMildMildModerateModerateSevereMildMildSeverity

Progressive

Slowly progressive

Slowly progressive

Slowly progressive

Slowly progressive

Progressive

StationaryStationary

Course

4431.54131.5Symptoms onset

15151555442.5CK increase XExon 7Exon 7Exon 7Exon 1Exon 6Exon 6Exon1Exon 1LMNA gene

DominantDominantDominantDominantSporadicSporadicSporadicDominant

Inheritance

MMFMFMMMSex42141225257158.6Age at study87654321Case no.

Base

SVC

Base

Apex

Figure 10.048-0.09±0.032-0.06±0.024Inf. Wall principal strain (%)NS55±1360±28RA vol/BSA (mls/m²)

NS38±641±25LA vol/BSA (mls/m²)NS0.63±80.58±11RVEF (%)NS69±1062±18RVEDV/ BSA (mls/m²)NS61±1460±18LV Mass/BSA (g/ m²)NS0.72±40.72±5LVEF (%)NS72±773±15LVEDV/ BSA (mls/m²)

P valueControlsPatientsParameter

NS3.5±1.073.4±1.14Gradient systole (S-1)0.027.13±2.74±1.2Gradient early diastole (S-1)NS2.4±0.644.6±2.7Ea/AaNS5±1.75±0.67E/EaNS0.08±0.040.05±0.02Pulsed TDI vel PW A (cm/s)NS0.19±0.060.18±0.05Pulsed TDI vel PW E (cm/s)NS0.11±0.040.13±0.02Pulsed TDI vel PW systole (cm/s)NS1.69±0.51.73±0.3Mitral E/A

NS63.3±1263.3±12LV Mass/BSA (g/ m²)

NS38±835±2LV ∆D %P valueControlsPatientsParameter

Table 1

Table 2

Table 3

Echo result are described in Table 3. Patients had normal left ventricular size and function and there was no significant difference between patients and controls. Mitral E and A wave diastolic velocities were similar for both groups as were the annular E to A wave ratios Pulse tissue Doppler velocities were normal in all mitral annular regions and were similar between patients and controls. The mitral E to annular E’ wave ratios was also normal Systolic myocardial velocity gradients were similar between patients and controls. The early diastolic gradient was lower however in patients than in controls suggesting early myocardial dysfunction (4 ± 1.2 s-1 vs. 7.1 ± 2.7 s-1, p=0.02)

Conclusions: This study suggests that unlike Duchenne or Becker’s muscular dystrophy, cardiac involvement in EDMD does not appear to involve myocardial fibrosis in the early stage of the disease. While global systolic function may appear normal using conventional echocardiography and CMR cine imaging, there appear to be subtle myocardial abnormalities manifested as a reduction in early diastolic myocardial gradients by echo and systolic strain patterns by CMR. Routine measurement of these may be useful in the early detection of cardiac involvement in EDMD.