Cardiac involvement in patients with Becker muscular dystrophy: new diagnostic and pathophysiological insights by a CMR approach

In a prospective two-center-study, 15 patients with BMD were consecutively ascertained and studied with respect to their neurological disorder at the neuromuscular outpatient clinic of the Department of Neurology, University of Ulm, Germany, followed by a comprehensive cardiological work-up at the Division of Cardiology of the Robert-Bosch-Krankenhaus Stuttgart, Germany. The diagnosis of BMD was confirmed by skeletal muscle biopsy evaluation with immunohistochemical analysis of the dystrophin protein and/or a genetic work-up with screening for dystrophin gene deletion and duplication. Cardiological work-up included physical examination, 12-lead ECG, echocardiography and CMR. Blood samples were obtained for laboratory studies including measurement of total creatine kinase (CK), cardiac troponin-I (TnI), brain-natriuretic-peptide (BNP) and C-reactive-protein (CRP) apart from others. This study was approved by the local ethical committee. Written informed consent was obtained from each patient or his legal guardian before inclusion to this study.

Two-dimensional echocardiography was performed on a commercially available system (Vivid 7, GE Healthcare) using a 3.5-MHz transducer with the patient in the left lateral decubitus position. Standard two-dimensional long and short axis views were recorded with the transducer in the apical and parasternal position. Measurements included left ventricular (LV) endsystolic and enddiastolic volumes. Biplane LV ejection fraction (LVEF) was calculated using the Simpson's rule. A semiquantitative global wall motion score index (GWMS) was calculated based on a 16-segment model as recommended by the American Society of Echocardiography [12]: two experienced echocardiographers blinded to the patient's clinical information visually scored wall motion on a 0 to 3 scale for each wall segment (normal = 0, hypokinetic = 1, akinetic = 2, dyskinetic = 3). The GMWS was then calculated as the sum of the myocardial wall motion segment scores divided by the number of segments assessed.

Electrocardiographic-gated CMR was performed in breath-hold using a 1.5-T system (Magnetom Sonata, Siemens Medical Solutions, Erlangen, Germany). Both cine and late gadolinium enhancement (LGE) short-axis CMR images were prescribed every 10 mm (slice thickness 6 mm) from base to apex. In-plane resolution was typically 1.2 × 1.8 mm. In addition to cine CMR, LGE images were acquired on average five to ten minutes after contrast administration with the use of a segmented inversion recovery gradient echo technique with constant adjustment of inversion time [13]. The contrast dose (Magnevist^®, Bayer-Schering, Germany) was 0.15 mmol/kg. In addition, fat-saturated and T2-weighted images were obtained to allow differentiation among subepicardial LGE, epicardial fat, and pericardial effusion.

Cine and contrast images were evaluated by two blinded observers as described elsewhere [14]. In brief, endocardial and epicardial borders were outlined on the short-axis cine images. Volumes and EF were derived by summation of epicardial and endocardial contours. The extent of LGE was planimetered applying ImageJ software (National Institutes of Health, Bethesda, Md, USA) on the short-axis contrast images with the use of an image intensity level ≥4 SD above the mean of remote myocardium to define LGE indicative of myocardial damage. In addition, a semiquantitative GWMS was calculated based on a 16-segment model with cine short-axis CMR images similar to echocardiographic analysis.

Data for continuous variables are expressed as median values in addition to minimal and maximal values, whereas data for categorical variables are expressed as the number and the percentage of patients. Comparisons between groups were done using a two-sided unpaired Student's t-test for normally distributed, and Mann-Whitney U test for non-normally distributed continuous variables. For categorical variables we used the chi-square test and Fisher's exact test, where appropriate. Agreement between methods was measured with Kappa statistics. Simple linear regression analyses were performed in order to identify possible determinants of reduced LVEF. The difference between two groups was defined to be statistically significant if the two-sided p-value was < 0.05.

The patient characteristics are shown in Table 1. The median age of this collective was 37 years with a maximum of 56 years and a minimum of 11 years. Muscle biopsy had been performed in 13 patients with 12 of them demonstrating reduced levels of dystrophin in the immunohistochemical stainings. Genetic analyses for dystrophin gene deletion and duplication mutations were available in 12 patients with eight of them demonstrating DNA mutations. Cardiac involvement indicating early or advanced DCM had been previously diagnosed in seven patients by echocardiography.

As expected, all patients had elevated serum CK enzymes, however, all but one patient were TnI-negative suggesting a skeletal muscle origin of these enzymes (Table 2). Interestingly, BNP serum values were elevated only in two patients in spite of depressed LV function (LVEF <60%) in 10 patients as demonstrated by CMR analyses. ECG abnormalities were observed in seven patients (Table 2). However, only four of these seven patients demonstrated those ECG abnormalities previously described to be typical in patients with MD [15] including a R:S-ratio ≥1.0 in lead V1, a deep Q-wave in leads I, II, aVL, V5-V6 or a complete right bundle branch block (Fig. 1).

Abnormal echocardiographic results defined as LVEF<60% and/or visual observation of rWMA were present in eight of 15 (53.3%) patients with all eight patients demonstrating reduced LVEF as well as some rWMA (median GWMS = 0.19; range 0 to 1.19). Interestingly, CMR revealed abnormal findings (defined as LVEF<60% and/or rWMA and/or presence of LGE) in 12 of 15 (80.0%) patients (p = 0.040 compared to echocardiography) with 10 (66.6%) of them having reduced LVEF (median LVEF = 52%; range 27% to 70%; p = 0.16 compared to echocardiography) as well as 9 (64.3%) demonstrating rWMA (median GWMS = 0.19; range 0 to 1.06; p = 0.38 compared to echocardiography). Both techniques revealed rWMA in the same eight patients. However, using echocardiography rWMA was mostly observed in septal wall segments (n = 4) followed by lateral segments (n = 2) while CMR revealed rWMA primarily in the (infero-)lateral segments (n = 6). The kappa-test showed good agreement in the assessment of rWMA with echocardiography compared to CMR when studies without any rWMA were included to the analysis (kappa = 0.86). However, there was only a modest agreement in the assessment of rWMA when studies without rWMA were excluded (kappa = 0.37).

Myocardial damage as indicated by LGE imaging was detected in 11 of 15 (73.3%) patients with a median myocardial damage extent of 13.0% (range 0 to 38.0%; Fig. 2). The most common pattern of LGE was a subepicardial to intramural extension in the inferolateral wall (n = 8) followed by an intramural location in septal segments (n = 2) and a transmural pattern in the lateral wall in one patient. A substantial agreement between the presence of myocardial damage as demonstrated by LGE imaging and rWMA was detected by kappa analysis (kappa = 0.69). Interestingly, two patients with preserved systolic LV function also demonstrated subepicardial LGE in the inferolateral wall. Correlation and regression analyses revealed a substantial association between increasing myocardial damage extent measured by LGE imaging and higher age (r² = 0.770; p < 0.001). The youngest patient demonstrating at least some LGE was found to be 26 years of age and all patients older than 26 years had some myocardial damage. LV systolic function was depressed in the absence of LGE only in one patient aged 23 years.

When patients with normal LVEF (≥60%; n = 5) were compared to those with reduced LVEF (<60%; n = 10), significant differences were detected for age, BMI, LV-EDV, LV-mass, LVEDD, GWMS and LGE extent (Table 3). Right ventricular systolic function tended to be also reduced in those patients with depressed LV function, however, this difference did not reach statistical significance. In order to find out determinants of reduced LVEF (as measured by CMR) we performed simple linear regression analysis (Table 4 and Fig. 3): There was a significant decline of LVEF with increasing myocardial damage extent (as measured by LGE imaging). Decreased LVEF was also associated with increasing LV-EDV, LV mass and LVEDD. As expected, the highest coefficient of determination of LVEF was calculated for the parameter GWMS confirming that reduction in LVEF is due to the extent and severity of rWMA. Furthermore, higher age was associated with a decrease in systolic LV function. A meaningful multivariate regression analysis could not be performed due to the limited size of this cohort.

Clinical examination revealed skeletal muscle involvement in 12 patients (80.0%), however, at least minor degrees of skeletal muscle weakness and pathology were detected by EMG in all patients limiting their effort and exercise capacity, thereby potentially masking cardiac exercise-dependent symptoms. Only four patients (26.7%) were already taking heart failure medication (Table 1) although cardiac involvement including early or advanced DCM had been previously diagnosed in seven patients (46.7%). CMR examination revealed cardiac involvement in five more patients increasing the number of patients with abnormal left ventricles to 12 (80.0%; p = 0.019) with 10 of them (66.7%) having reduced LV systolic function. Therefore, these patients were regarded as in need for medical therapy (p = 0.009 compared to those patients already on therapy) considering the recommendations of the guidelines on heart failure therapy [9].

Cardiac involvement has been described in patients with MD for decades by post-mortem analyses [15, 16]. In previous studies, mainly echocardiographic and scintigraphic methods [4, 15, 17] were applied for cardiac evaluation of patients with MD: In BMD patients, cardiac involvement was found to be uncommon in patients aged <16 years, while more than 70% did demonstrate pathologic cardiac findings by the age of 40 years [4, 17]. Our results are in accordance with these previous studies, since the only three patients without any pathological finding were aged 11 years to 14 years. However, this study extends previous knowledge, since we could demonstrate pathological findings (including reduced LVEF and/or presence of LGE) by CMR in all patients aged ≥23 years (n = 12; 80.0%) while conventional echocardiography revealed pathological results only in eight patients aged ≥23 years (53.3%; p = 0.040). The main advantage of CMR as compared to conventional echocardiography is the possibility to detect myocardial damage resulting in abnormal examinations in 12 patients.

Blood sample analysis for total CK revealed increased CK levels in all patients included to this study. Interestingly, higher CK-values were found in those with still preserved LV function. This finding is in accordance with previous studies suggesting that there is no correlation between cardiac disease severity and the degree of skeletal myopathy [4, 17]. Surprisingly, elevated BNP levels were observed only in two of the ten patients with depressed LV function. One explanation for this observation may be that the degree of LV dysfunction is not severe enough to cause a rise in BNP [18]. Since BNP levels are supposed to also reflect rapid changes of pressure gradients in the ventricles [19, 20], one may argue that the underlying pathophysiology in muscular dystrophy is mainly driven by continuous myocardial cell death (due to dystrophin absence or fragility) resulting secondarily in a continuous and adapted ventricular enlargement with progressive decrease of systolic function without quick changes in ventricular filling pressures.

In the past, ECG changes attracted great interest as sensitive signs of cardiac involvement in BMD patients. The typical ECG changes described in the literature were felt to reflect a reduction in electromotive forces in the posterobasal and contiguous lateral wall [15, 21]. However, the shortcomings of a cardiac examination relying mainly on ECG findings are illustrated in this study: Only seven patients (46.7%) did demonstrate some ECG abnormalities with only four patients (26.7%) having those ECG changes that were described previously to be typical in muscular dystrophy. In comparison to ECG, 12 patients (80.0%) had evidence of cardiac involvement by CMR analysis.

The comparison of patients with normal and reduced LVEF confirmed previous results: patients with reduced LVEF were older, had larger and heavier hearts and demonstrated more rWMA. Left ventricular characteristics such as LV-EDV, LV mass and LVEDD were significantly associated with LVEF as shown by simple linear regression analysis. Interestingly, we found a substantial agreement between the presence of myocardial damage as demonstrated by LGE imaging and rWMA (kappa = 0.69). Since the highest coefficient of determination of LVEF was calculated for the parameter GWMS which depends on the extent and severity of rWMA which in turn is related to regionally pronounced myocardial damage, we conclude that the progressive reduction in LVEF is primarily due to the extent of myocardial damage.

In post mortem hearts, it has already been demonstrated that myocardial damage is starting from the epicardial third of the inferolateral wall with possible transmural extension in contiguous segments [16, 22] and a genetical predilection of the posterobasal and lateral LV walls in patients with MD was suggested [21]. Our results obtained in the present study by state-of-the-art non-invasive CMR imaging confirm those previous results and impressively illustrate the tremendous progress in medicine: assessment of myocardial damage which could only be done by invasive ex vivo histopathological work-up at necropsy in the 1960s and 1970s, is today achieved by non-invasive in vivo CMR. Recently, Silva et al. reported their CMR findings in a small and considerably different collective (compared to ours) of only 10 young patients primarily suffering from DMD. They observed myocardial fibrosis in seven out of their 10 patients with the lateral wall being most commonly involved [23].

The detailed cellular and molecular mechanisms leading to myocardial contractile dysfunction in patients with MD are still to be elucidated. Preliminary animal studies suggest early alterations in cell metabolism and signal transduction in dystrophin-deficient hearts with a possible defect in the NO/cGMP pathway [24] and excessive intracellular Ca-signaling and ROS generation with breakdown of the mitochondrial membrane potential [25]. Interestingly, changes in regional myocardial metabolism have been documented at early disease stages, when rWMA or fibrosis were still absent [21]. Therefore, morphological changes in the myocardium of patients with MD are believed to be a consequence of the underlying genetically-determined metabolic and structural abnormalities which can be aggravated by additional mechanical stress.

Interestingly, our group as well as other research groups have shown previously that this subepicardial pattern of myocardial damage which was detected predominantly in the posterolateral myocardial wall of patients with BMD is also a characteristic finding in biopsy-proven myocarditis [14, 26, 27]. This typical feature has become useful for non-invasive diagnosis of myocarditis by CMR (Fig. 4). Indeed, a final common pathway leading to myocardial damage in patients with MD and (entero-) viral myocarditis has been suggested previously [28, 29]. In both diseases, progressive LV dysfunction is thought to be the result of dystrophin protein disruption [30‐32]. Although cleavage of dystrophin has only been documented for coxsackievirus B3 [30, 33], similar mechanisms for induction of myocardial damage can be envisaged also for other cardiotropic viruses. However, this needs further studies.

As we did not perform endomyocardial biopsies in our patients with BMD, we could not rule out that myocarditis was the ultimate cause of the observed posterolateral pattern of myocardial damage. However, due to the fact that almost all patients aged ≥23 years showed this pattern, and all patients had proof of genetically determined abnormalities of dystrophin protein, an additional infectious cause seems highly unlikely.

How can one explain that a presumably diffusely distributed dystrophin abnormality in the human heart results in a pattern of myocardial damage which is preferentially located in the subepicardial portions of the posterolateral wall? One possibility is that the genetically determined dystrophin damage is not diffusely distributed but accentuated in the posterolateral wall [16]. The other possibility is that a diffuse (genetically determined) disease process with ubiquitous alterations in cell metabolism and signal transduction precedes functional impairment [24, 25] leading to myocardial damage preferentially in the posterolateral wall. Based on this hypothesis, the myocardial damage in the posterolateral wall could represent a non-specific cardiac phenotype in response to exaggerated mechanical stress in this region. The fact that a diversity of presumably diffuse diseases such as sarcoidosis, Chagas disease, Anderson-Fabry disease [34] or arrhythmogenic right ventricular cardiomyopathy [35] demonstrate the main damage in the posterolateral wall makes this yet unproven hypothesis attractive. Different pathophysiologies (such as myocardial inflammation or intracellular accumulation of storage products) may predispose the human heart to myocardial damage, which is amplified by increased mechanical stress and characterised by necrosis and/or fibrosis beginning in the inferolateral wall. Whether the inferolateral wall constitutes the "weakest cadet" of LV segments due to metabolic, structural or functional characteristics or otherwise has to withstand the highest mechanical stress during LV contraction and relaxation still has to be explored.

Early cardiac evaluation including state-of-the-art non-invasive imaging of young patients with MD is required in order to identify those who will most likely benefit from early heart failure therapy. Timely onset of heart failure medication has been shown to result in beneficial ventricular remodelling with improvement in LV systolic function or at least in retardation of progressive cardiac dysfunction [6, 7, 36]. In our study group, heart disease had already been established by echocardiographic studies in seven patients, however, only four (26.7%) were taking heart failure medication. Obviously, the decision about putting a patient with MD on heart failure medication cannot only be based on clinical symptoms, since most patients suffer from skeletal myopathy and therefore avoid exaggerated physical activity, thereby masking potential symptoms. When considering the potential rapid progression of cardiac involvement in these patients and the proven beneficial effects of medical therapy [6, 7, 36], extensive and regular cardiac evaluations using CMR should become the standard of cardiac care for these patients. Recently published ACC/AHA guidelines on the diagnosis and treatment of heart failure suggest that heart failure therapy should be begun even in the absence of clinical symptoms if there is evidence of structural abnormalities and reduced LV function [8, 9]. Compared to this standard, the majority of our patients were hitherto undertreated. We started medical heart failure therapy in those patients in this study who had a decreased LVEF (n = 10; 66.7%) independent of their symptoms. Whether patients who only show some amounts of LGE in spite of normal LVEF – as was the case in two patients – will also benefit from early medical therapy, has to be explored in future studies.

The study size was small (with only 15 BMD patients) preventing further statistical analyses. However, considering the prevalence rate of this rare disease, it still constitutes a substantial number and no comparable CMR-based comprehensive studies focusing on LGE-imaging have been performed on a cohort of BMD patients previously, although there is a prior case report [37]. We acknowledge that new techniques such as CMR-based strain imaging [38] or echocardiography-based tissue-doppler and strain imaging [39] represent promising tools for early detection of subclinical LV dysfunction in MD patients. However, these techniques are still under investigation and do not belong to the established and clinically used armentarium of most centers. Although the value of these methods in comparison to CMR-based LGE imaging will be of great interest, for the purpose of this study only well accepted and widely established echocardiographc and CMR-based methods were applied.