Late gadolinium enhanced cardiovascular magnetic resonance of lamin A/C gene mutation related dilated cardiomyopathy

Dilated cardiomyopathy (DCM) is a major cause of heart failure and sudden cardiac death. About one third of DCM cases are familial. During the past few decades, several DCM disease genes have been identified, many of them limited to individuals or families [1‐3]. The lamin A/C gene (LMNA) is so far the most significant disease gene for DCM. It has been estimated that LMNA mutations cause up to 10% of familial DCM. The penetrance of the LMNA mutations causing cardiomyopathy is nearly complete [4]. The gene resides on chromosome 1q21.2-q21.3. By alternative splicing, it codes for lamins A and C, proteins found in the nuclear lamina. LMNA mutations may also cause a variety of other clinical entities such as neuromuscular syndromes, lipodystrophy or premature aging disease, i.e. progeria. We have previously described several LMNA mutations among Finnish DCM patients [2, 3].

Carriers of LMNA mutations causing cardiomyopathy should be recognised at an early stage so that regular cardiologic follow-up can be arranged; affected individuals have a high risk of sudden cardiac death or severe heart failure by middle age [5]. However, most of the patients are asymptomatic or only have mild symptoms for many years, and diagnosis is therefore difficult. Moreover, the serious nature of LMNA-related cardiomyopathy is easily overlooked at the early stage because the left ventricular (LV) dilatation is typically mild at the beginning, and it may remain so even later on although the LV ejection fraction may be very low. In other words, LMNA mutations may cause DCM, or cardiomyopathy with decreased systolic function, combined with LV dilatation that is too mild to fulfil the criteria of DCM. The earliest cardiac finding in cardiomyopathy caused by a LMNA mutation is usually conduction system disease, associated with atrial or ventricular arrhythmias, atrioventricular (AV) conduction abnormalities and myocardial dysfunction [6].

Echocardiography is regarded as the standard method to clinically study DCM; however, cardiovascular magnetic resonance (CMR) plays an important role in the characterisation and risk stratification of patients with DCM. Late gadolinium enhancement (LGE) CMR with standard cine-imaging is an accurate and reproducible method both to demonstrate myocardial fibrosis and to measure wall thickness as well as global and regional function of both ventricles [7, 8].

The main purpose of this study was to characterise myocardial fibrosis, regional wall motion abnormalities, ventricular dilatation, longitudinal LV systolic function and global function with LGE CMR in asymptomatic or mildly symptomatic carriers of LMNA mutations causing DCM. In addition, we investigated the possible association between the localisation of myocardial fibrosis and the conduction abnormalities documented with electrocardiography (ECG).

Out of 17 patients, 15 exhibited LGE. Among the total of 289 segments, LGE was observed in 47 (16%). The extent of LGE was less than 50% in 36/47 enhanced segments (Figure 1). In all patients, LGE occurred in the basal or mid-ventricular septal wall (Figure 2, 3). RV free wall enhancement was not observed. Thirty three of the 47 enhanced segments were in the septal area (segments 2-3, 8-9, 14). The typical pattern of LGE was linear and mid-myocardial (Figure 4). Diffuse enhancement of LV myocardium was observed in two advanced cases of the disease. LV wall thickness was normal in all patients. Abnormal motion was found in 27 segments (9%). The motion abnormalities were predominantly located in the basal regions of LV (segments 1-4). There was hypokinesia in 17 segments and akinesia in 10 segments. The correlation between the extent of enhancement and abnormal motion was significant (p < 0.001). Abnormal motion was found in 50% of segments enhancing 26-50% and in all segments if the enhancement was over 50% (Table 2).

All patients with conduction abnormalities (11 with an AV conduction defect and two with a widened QRS complex) and two with atrial fibrillation showed enhancement of the basal and mid-ventricular septum. Conduction abnormalities correlated significantly with presence of LGE in the LV septum (p = 0.01). Four patients had a normal ECG recording, and two of them had enhancement of the basal septum. Table 3 summarises the ECG abnormalities in relation to the enhanced segments and regional wall motion abnormalities.

A volumetric analysis of one female subject was excluded because of poor cine image quality due to arrhythmia. When compared with the age, sex and BSA standardised normal values published earlier by Maceira et al [8, 9], ventricular volumes and systolic function were preserved in five patients. Minor systolic dysfunction without ventricular dilatation was observed in three patients (LV dysfunction in one patient and dysfunction of both ventricles in two patients). Mild LV dilatation without systolic dysfunction was observed in four patients, two of whom also had a slightly dilated RV. A dilated LV and systolic dysfunction was observed in four patients, three of whom also had RV dysfunction, and one female subject had RV dilatation without dysfunction. Table 4 summarises the global ventricular functions and ejection fractions for each of the 16 patients.

Longitudinal LV systolic function was decreased in nine patients when compared to normal values published by Nikitin et al [13]. Decreased MAD was observed in eight patients with decreased EF, three patients (18%) with normal EF showed longitudinal LV systolic dysfunction (Table 3).

MAD was significantly lowered when EF was decreased (p < 0.03) (Figure 5). MAD correlated significantly with percentage of myocardial LGE (p < 0.01) (Figure 6).

The present study demonstrates that 88% of asymptomatic or mildly symptomatic carriers of LMNA mutations causing cardiomyopathy had typical myocardial fibrosis, predominantly in the mid-myocardium of the basal septum. This was observed in all individuals with an AV conduction defect. The pattern of enhancement was typically linear and less than 50% of the area of the segment. Enhancement was associated with wall motion abnormalities at LV basal segments, where the degree of enhancement had a significant correlation with decreased motion. Minor systolic dysfunction without ventricular dilatation, preserved systolic function with mild LV dilatation or systolic dysfunction with mild dilatation of LV, or of both ventricles, was observed in 69% of LMNA mutation carriers. In our patients as an early marker of the disease longitudinal LV systolic function was decreased in nine patients when compared to normal values [13]. In addition, we found that LV EF as well as percentage of myocardial fibrosis correlated significantly with abnormal MAD.

Early studies show that even a very small amount of myocardial scar tissue is visible with routine LGE CMR sequences [14]. Myocardial diseases have typical enhancement patterns on the basis of their characteristic pathology [15]. In previous studies, about one third of DCM patients have demonstrated mid-myocardial linear enhancement on LGE CMR in a non-coronary distribution, predominantly in the LV basal and mid-ventricular septum [8, 16‐19]. However, Ebeling Barbier et al have documented that 30% of elderly volunteers without or with minimal coronary artery disease showed myocardial LGE, but mostly inferolateral region of the LV. On the other hand, Breuckmann et al found that 12% of marathon runners had LGE. Subendocardial and transmural LGE pattern was found in five marathon runners, but also non-CAD pattern was documented in seven subjects [20]. A healthy control group of DCM patients in study by Zimmerman et al [18] showed no myocardial LGE. As LGE imaging is widely used as a gold standard in cardiac research, we did not collect any new data of healthy subjects [21]. In our study, 88% of LMNA mutation carriers had typical myocardial fibrosis, predominantly mid-myocardially in the basal septum. High degree of LGE in our patients may reflect the effect of these mutations alone and not the effects of Lamin A/C mutations in general.

Assomull et al reported that mid-wall myocardial fibrosis in DCM patients predicted both ventricular tachycardia and sudden cardiac death [17]. Cardiomyopathy caused by a LMNA mutation has a malignant course and affected individuals are at high risk of sudden cardiac death [6]. The earliest cardiac finding is usually conduction system disease [22]. Raman et al demonstrated mid-myocardial fibrosis and abnormal LV diastolic function with LGE CMR in patients with DCM caused by a LMNA mutation [23]. This is in line with our findings, the majority of our patients having myocardial scarring and regional motion abnormalities in corresponding areas.

Previous autopsy studies have shown the location of fibrosis near the region of the conduction system and this has been postulated to be the cause of the AV blocks in LMNA-related cardiomyopathies [24, 25]. In consensus, cardiac conduction defects such as a prolonged PR interval, a widened QRS complex and atrial fibrillation correlated with the presence of septal wall fibrosis and motion abnormalities in our study.

Zimmermann et al concluded that enhancement is a typical feature on CMR carried out in DCM patients but it did not correlate with myocardial biopsy findings [18]. Since a biopsy only yields information about the selected biopsy site the detection of fibrosis with biopsy can be unsatisfactory due to the mid-myocardial location of fibrosis. Whereas LGE imaging can be used to reliably detect the location and extent of fibrosis, and it can even be used to offer guidance in biopsy sampling [26]. Sixty-nine percent of our asymptomatic or mildly symptomatic patients also showed mild ventricular dilatation and systolic failure in global analysis. Twelve of our 17 patients were also participants in the study of Koikkalainen et al, which determined an amalgam of parameters calculated from cine images typical for LMNA mutation carriers [27]. The study also found a significant difference between the patient and control group in volumetric and function parameters even in the early and asymptomatic stage of the disease. Fifty-three percent of our patients had decreased longitudinal LV systolic function. LV myocardial contraction and relaxation are reported to impair first in longitudinal direction in patients with heart failure [28, 29].

In conclusion, in the early stage of DCM caused by a LMNA gene mutation, AV conduction defects are common with or without a minor depression in LV systolic function. Usually the suspicion of DCM caused by LMNA mutations is based on the presence of conduction system disturbances in an ECG and by clinical exclusion of other aetiologies. However, other inflammatory or infiltrative cardiomyopathies may resemble DCM caused by LMNA mutations. If the typical pattern and location of myocardial fibrosis, mild LV dilatation and systolic ventricular dysfunction can be accurately observed with LGE CMR in the early stage of the disease, the diagnosis can be confirmed by DNA analyses. Myocardial scarring is known to provide a substrate for ventricular dysfunction, conduction abnormalities and ventricular arrhythmias. The disease is known to have a malignant course, and the combination of ECG recordings and LGE CMR will help to recognise these patients. LGE CMR can also be used to differentiate between LMNA-related cardiomyopathy and cardiac sarcoidosis, both of which can cause AV conduction defects and arrhythmias. Although no specific medical treatment exits so far to prevent the development of LMNA-related cardiomyopathy, sudden cardiac death may be prevented by a timely placement of a pacemaker or an implantable cardioverter defibrillator.

LGE CMR is also a reproducible method to monitor the progression of the disease and therefore we have an ongoing follow-up study with DE-MRI to obtain further information about the progression of the familial form of DCM caused by a LMNA mutation.