ACE indicates angiotensin-converting enzyme; BMD, Becker muscular dystrophy; D, dysfunction; DMD, Duchenne muscular dystrophy; ND-NF, no dysfunction–no fibrosis; ND-NT, no dysfunction–no treatment; and ND-T, no dysfunction–treatment.
A, Progression of MF from baseline to the 2-year follow-up cardiovascular magnetic resonance (CMR) examination for all patient groups. B, Significant inverse correlation of baseline left ventricular ejection fraction (LVEF) and baseline MF identified on CMR. C, Significantly higher probability of cardiovascular events in patients with MF compared with those without MF. D indicates dysfunction; ND, no dysfunction; NF, no fibrosis; and NT, no treatment.
A. Late gadolinium-enhancement left ventricular (LV) short-axis section images of a patient with Duchenne muscular dystrophy (DMD) and MF. Red arrowheads indicate the inferolateral subepicardial and midwall contiguous fibrosis; yellow arrowheads, the anterior segment with contiguous MF; and blue arrowheads, multifocal anteroseptal fibrosis. B, Late gadolinium-enhancement LV long-axis section images of the same patient with DMD and MF. Note the correlation between inferolateral segments MF (red arrowhead) in short-axis, long-axis, and anterior segments (yellow arrowheads). C, Patient with DMD from the nontreated group (no dysfunction–no treatment) at baseline cardiovascular magnetic resonance (CMR) with little MF at this level and a small amount of MF for the entire LV. D, Same patient at follow-up CMR with important progression of MF at the same level and for the whole LV (from 1.7% to 9.1% of LV mass). E, One short-axis section of 2 patients with DMD (1 in each row) with typical MF pattern seen in our group of patients and with quantitative measurement examples using a threshold of signal intensity above 5 SDs of mean normal myocardium (depicted in yellow). Images in the first column have no region-of-interest drawings; those in the middle column show only the endocardial (red outline) and epicardial (green outline) LV contours for visual comparison.
eMethods. Data Analysis Plan
Customize your JAMA Network experience by selecting one or more topics from the list below.
Silva MC, Magalhães TA, Meira ZMA, Rassi CHRE, Andrade ACDS, Gutierrez PS, Azevedo CF, Gurgel-Giannetti J, Vainzof M, Zatz M, Kalil-Filho R, Rochitte CE. Myocardial Fibrosis Progression in Duchenne and Becker Muscular DystrophyA Randomized Clinical Trial. JAMA Cardiol. 2017;2(2):190–199. doi:10.1001/jamacardio.2016.4801
Copyright 2016 American Medical Association. All Rights Reserved.
Does early use of angiotensin-converting enzyme inhibitor therapy in patients with Duchenne or Becker muscular dystrophy who have preserved left ventricular ejection fraction affect the progression of myocardial fibrosis defined as percentage of left ventricular mass identified by cardiovascular magnetic resonance?
In this randomized clinical trial of 42 patients with myocardial fibrosis and preserved left ventricular ejection fraction, those receiving angiotensin-converting enzyme inhibitors demonstrated significantly lower myocardial fibrosis progression compared with those who were untreated. Early use of angiotensin-converting enzyme inhibitor therapy was independently associated with a lower rate of myocardial fibrosis progression.
Early angiotensin-converting enzyme inhibitor therapy, initiated when myocardial fibrosis is identified before a decrease in left ventricular ejection fraction, seems to benefit patients with Duchenne or Becker muscular dystrophy.
In Duchenne muscular dystrophy (DMD) and Becker muscular dystrophy (BMD), interventions reducing the progression of myocardial disease could affect survival.
To assess the effect of early angiotensin-converting enzyme (ACE) inhibitor therapy in patients with normal left ventricular function on the progression of myocardial fibrosis (MF) identified on cardiovascular magnetic resonance (CMR).
Design, Setting, and Participants
A randomized clinical trial conducted in 2 centers included 76 male patients with DMD or BMD undergoing 2 CMR studies with a 2-year interval for ventricular function and MF assessment. In a non–intent-to-treat trial, 42 patients with MF and normal left ventricular ejection fraction (LVEF) were randomized (1:1) to receive or not receive ACE inhibitor therapy. The study was conducted from June 26, 2009, to June 30, 2012. Data analysis was performed from June 30, 2013, to October 3, 2016.
Randomization (1:1) to receive or not receive ACE inhibitor therapy.
Main Outcomes and Measures
Primary outcome was MF progression from baseline to the 2-year CMR study.
Of the 76 male patients included in the study, 70 had DMD (92%) and 6 had BMD (8%); mean (SD) age at baseline was 13.1 (4.4) years. Myocardial fibrosis was present in 55 patients (72%) and LV systolic dysfunction was identified in 13 patients (24%). Myocardial fibrosis at baseline was an independent indicator of lower LVEF at follow-up (coefficient [SE], −0.16 [0.07]; P = .03). Among patients with MF and preserved LVEF (42 [55%]), those randomized (21 patients in each arm) to receive ACE inhibitors demonstrated slower MF progression compared with the untreated group (mean [SD] increase of 3.1% [7.4%] vs 10.0% [6.2%] as a percentage of LV mass; P = .001). In multivariate analysis, ACE inhibitor therapy was an independent indicator of decreased MF progression (coefficient [SE], −4.51 [2.11]; P = .04). Patients with MF noted on CMR had a higher probability of cardiovascular events (event rate, 10 of 55 [18.2%] vs 0 of 21 [0%]; log-rank P = .04).
Conclusions and Relevance
In this 2-year, follow-up, randomized clinical trial of patients with Duchenne or Becker muscular dystrophy whose LVEF was preserved and MF was present as determined on CMR, ACE inhibitor therapy was associated with significantly slower progression of MF. The presence of MF was associated with worse patient prognosis.
clinicaltrials.org Identifier: NCT02432885
Duchenne muscular dystrophy (DMD) and Becker muscular dystrophy (BMD) are genetically determined diseases caused by a mutation in the dystrophin gene located on chromosome X, locus Xp21. Dystrophin is located on the myocyte membrane and connects the cytoskeleton with the extracellular matrix,1 being responsible to maintain cell membrane stability during muscle contraction. With the absence (DMD) or decreased amount (BMD) of dystrophin, myocytes develop successive microruptures of the sarcolemma during contraction, with subsequent necrosis and replacement by fibrofatty tissue.2 Clinically, after a short period of apparently normal motor development, there is progressive, irreversible skeletal muscle degeneration, with consequent muscle weakness.3 Of the 2 disorders, DMD is more severe and 10 times more frequent than BMD.4
Myocardial involvement is frequent, currently representing one of the main causes of death as a result of an improvement in respiratory care in recent years.5,6 In a previous study, our group reported for what we believe to be the first time the presence of myocardial fibrosis (MF) by cardiovascular magnetic resonance (CMR) in patients with DMD or BMD.7 However, potential prognostic implications and pharmacologic interventions that might affect the progression of MF were not investigated. The aim of the present study was to investigate the course of cardiac involvement in patients with DMD or BMD using a comprehensive serial CMR assessment, testing the prespecified hypothesis that early intervention with angiotensin-converting enzyme (ACE) inhibitors can favorably change the development of MF (the primary outcome) in patients with no left ventricular (LV) systolic dysfunction through a prospective randomized clinical trial.
This prospective study, conducted from June 26, 2009, to June 30, 2012, included male patients (≥6 years) with confirmed DMD or BMD who were being monitored in the specialized outpatient clinic for DMD or BMD of Federal University of Minas Gerais, Belo Horizonte, or Heart Institute, InCor, University of São Paulo, São Paulo, both in Brazil. The diagnosis of DMD and BMD was confirmed by the presence of dystrophin gene mutation in molecular analysis by multiplex ligation-dependent probe amplification (MLPA; MRC-Holland)8 and/or the absence or reduction of dystrophin determined on muscle biopsy results. For patients with no detectable deletion/duplication in the dystrophin gene on DNA analysis (multiplex polymerase chain reaction screening, and/or MLPA study), muscle biopsy was mandatory, with specific immunostaining and quantitative analysis for the dystrophin protein. The genetic analysis was performed at the Human Genome and Stem Cell Center Biosciences Institute, University of São Paulo, São Paulo, Brazil. Among patients without a mutation shown on genetic testing who refused to undergo the muscle biopsy, only 2 patients with a family history of DMD or BMD (each had a sibling with the disease confirmed by muscle biopsy) were included in this study. The patients or a responsible adult signed an informed consent form and the study was approved by the institutional ethics committees (Federal University of Minas Gerais Institutional Review Board and Institutional Board Review for Research Project Analysis of the Hospital and the Faculty of Medicine of the University of São Paulo). The participants did not receive financial compensation. The study protocol is available in Supplement 1.
All patients underwent routine cardiac evaluation at baseline that included a complete clinical examination, electrocardiogram, chest radiography, determination of the serum creatine kinase (CK) level, and Doppler echocardiography in our specialized pediatric DMD/BMD center. After signing the written informed consent form, patients were scheduled to undergo the baseline CMR examination and clinical evaluation, which were performed on the same day.
All patients underwent 2 CMR studies using the same protocol (Magnetom Avanto, 1.5-T; Siemens), with 1 performed at baseline and 1 at 2-year follow-up, with cine magnetic resonance imaging (MRI) for the assessment of LV contractile function (global and segmental) and late gadolinium enhancement (LGE) imaging for the detection and quantification of MF. Cine MRI (gradient-echo sequence in steady-state free precession)9 and LGE (fast gradient-echo with a preparatory inversion-recovery pulse)10 were acquired using standard settings in the same LV short- and long-axis views, allowing accurate comparison between LV function and myocardial structure. The LGE images were acquired 10 to 20 minutes after intravenous injection of gadolinium-based contrast (gadoversetamide, 0.2 mmol/kg; Optimark) and adjustment of the inversion time to null the signal of normal myocardium.
Digital imaging and communications in medicine images were analyzed in a central core laboratory. Qualitative and quantitative analyses of cine MRI and LGE were performed blindly and independently: cine MRI by one of us (M.C.S.) without knowledge of the treatment received by patients and LGE by 2 others of us (C.H.R.E.R. and C.E.R.) as observers blinded to any other patient data. Interobserver and intraobserver variability for 20 patients of our population were calculated using absolute agreement intraclass correlation coefficients with 2-way mixed models, Bland-Altman analysis, and coefficient of variation.
Segmental LV contractility and LGE presence were evaluated in short-axis images, using the American Heart Association 17-segment model.11 Left ventricular end-diastolic volume, end-systolic volumes, LV mass, and LV ejection fraction (LVEF) were calculated from steady-state free precession cine MRI images using the Simpson method as previously described12 (Leonardo Workstation; Argus-Siemens Healthcare). The LGE images were analyzed using dedicated software (cmr42, version 3.4.2; Circle Cardiovascular Imaging) with a semiautomatic quantitative tool for MF measurements, based on a threshold technique (MF measured as any pixel with signal intensity >5 SDs of the mean of normal myocardium) (Figure 1). The observer visually defined the normal myocardium as an area of homogeneous nulled myocardium (low signal) and drew a region of interest used to calculate the mean (SD) of normal myocardium; inferoseptal was an often-used segment because of frequent lateral wall involvement in our patients. Software that was validated and approved by the US Food and Drug Administration identified the area with pixel values more than 5 SDs above the mean, which defined LGE, the surrogate for MF. To our knowledge, no definition of an ideal threshold for detecting MF in patients with DMD or BMD cardiomyopathy is available. Previous studies directly investigating the different techniques used for myocardial fibrosis quantification in nonischemic cardiomyopathy have demonstrated that full width at half maximum, 5 SDs, 6 SDs, and manual threshold performed similarly and provided robust reproducibility.13-15 Based on previous data7 and a pilot analysis that we performed, the 5-SD cutoff above normal myocardial signal intensity had the best agreement with visual LGE analysis in a range of DMD and BMD cardiomyopathy severity.
Despite high-quality, no-artifact images in almost all cases, it was unavoidable that a few of the younger patients with spine deformities had difficulties remaining within the magnet and performing appropriate breath holding, leading to image artifacts promptly recognized by the expert observer. Clear and definitive artifacts (based on 2 expert decisions) were manually excluded with a special tool provided by the postprocessing software; therefore, those pixels were not counted as true LGE. A true LGE was also confirmed in both views of LV short axes and LV long axes images.
Patients were classified into 3 groups based on the presence of LV dysfunction and/or MF. The first group consisted of patients with LVEF less than 50% by echocardiography or CMR (group D [dysfunction]). The second group consisted of patients with LVEF of 50% or more and without MF by CMR (group ND-NF [no dysfunction–no fibrosis]). The third group consisted of patients with LVEF of 50% or more and with MF by CMR who were included in the non–intent-to-treat randomized trial.
Group D patients received conventional care for DMD and BMD and were treated with medications for systolic dysfunction at the discretion of the cardiologist. The participants received ACE inhibitors (enalapril, 10-20 mg, every 12 hours), β-blockers (a total carvedilol dose of 1 mg/kg, starting at 3.125 mg every 12 hours to reach the maximum dose of 25 mg every 12 hours), and in patients with LVEF less than 35%, an aldosterone inhibitor (spironolactone, 25 mg once daily). Group ND-NF patients received conventional care for DMD and BMD and did not receive any treatment for cardiomyopathy or heart failure.
Patients with LVEF of 50% or more and the presence of MF noted on CMR were randomized into 2 subgroups (block randomization 1:1, block size 4, performed by C.F.A.). In the first subgroup (group ND-NT [no dysfunction–no treatment]), patients received all necessary treatments according to their clinical condition based on guidelines16 (eg, corticosteroids, physical therapy, and respiratory care), but no treatment for cardiomyopathy or heart failure (ie, no treatment with ACE inhibitors, angiotensin receptor blockers, β-blockers, diuretics, or aldosterone inhibitors). The second subgroup (group ND-T [no dysfunction–treatment]), in addition to the same basic therapy applied to group ND-NT, received an ACE inhibitor (enalapril). No other medications were used for cardiomyopathy or heart failure in this group. The enalapril dose was titrated to the patient’s maximum tolerated level or a maximum dose. Enalapril was administered at a starting dose of 2.5 mg every 12 hours in patients younger than 10 years and increased up to 5 mg every 12 hours. In patients 10 to 15 years, the initial dose was 5 mg every 12 hours, with a maximum dose of 10 mg every 12 hours. In patients older than 15 years, the dose was 10 to 20 mg every 12 hours. The investigator responsible for the drug treatment (Z.M.A.M.) was not blinded to the group allocation because she needed to apply the treatment regimen, but she was blinded to all CMR data, including the amount of MF. Patients receiving corticosteroid therapy were given an intermittent regimen: 0.75 mg/kg/d for 10 days intercalated with no corticosteroids for 10 days in successive cycles.
Epidemiologic methods for follow-up included ascertainment of events blinded to randomization status or CMR results. During the study period, patients were followed up by a clinical appointment at the specialized DMD/BMD outpatient clinic with an interval of 6 months or at any time if they had new symptoms. Clinical information was obtained mainly from the outpatient clinic visits and, if necessary, telephone interviews and medical records. The primary outcome of the study was the degree of MF accumulation between baseline and follow-up CMRs. Major cardiovascular events (ie, cardiovascular death, hospitalization for heart failure, cardiac arrhythmia requiring hospitalization, myocardial infarction, and fatal or nonfatal stroke) were monitored during clinical follow-up and adjudicated by 2 cardiologists (M.C.S. and T.A.M.) blinded to the research randomization.
Descriptive results were given as frequencies and percentages for categorical variables and as mean (SD) for continuous variables. Normality was checked by the skewness-kurtosis test.
Mann-Whitney test (difference of continuous variables according to the comparison groups), Pearson χ2 test (association between categorical independent variables), McNemar (proportions comparison), and Wilcoxon signed-rank test (means) for paired collected data from the baseline and follow-up CMR examination were used. Further statistical tests are presented in the eMethods in Supplement 2.
The association between continuous variables was tested by Spearman correlation coefficient. To evaluate the indicators of change of LVEF and MF, we used multivariate logistic and linear regression. The diagnostic performance of traditional tests (chest radiography, electrocardiogram, and echocardiogram) considered at least 1 test with abnormal results and compared it with the presence of MF detected by CMR.
The Kaplan-Meier method was used in survival analysis. Baseline comparisons of CMR variables had 76 observations, and follow-up CMR included 74 observations (2 patients died before the CMR follow-up examination). Sample size calculation was limited considering that the primary outcome (progression of MF determined with CMR) had not been previously investigated at the time of the study design. To estimate differences in the progression of MF, we used surrogates of myocardial injury previously reported for patients with DMD or BMD in treatment trials, such as differences in fractional shortening (36%)17 and LVEF (13%)18 by echocardiography and in T2 myocardial signal (29.7%)19 in a CMR study. In a superiority trial analysis for the treatment group and estimating mean MF progression of 15% for the nontreated group and 10% for the treated group (difference of 5%), with an SD for the measurement of 4.5%, power of 90%, and α level of 5%, the calculated sample size for each group was 18 patients (plus 15%, with 21 patients to account for eventual patient loss during the 2-year follow-up).
Analyses were performed using Stata, version 10.0 (StataCorp LP). Statistical significance was defined as 2-sided P < .05. Data analysis was performed from June 30, 2013, to October 3, 2016.
Between June 26, 2009, and June 30, 2012, a total of 76 male patients were included in the study: 70 patients with DMD and 6 with BMD (Figure 1). Mean (SD) age was 13.1 (4.4) years. Forty-three patients (57%) had muscle biopsy performed and 40 patients (53%) had identified DNA mutation. Mean interval between CMR studies was 2.05 (0.11) years (range, 1.9-2.5 years). Table 1 summarizes the characteristics of the study cohort, and Table 2 reports the progression of MF by CMR between baseline and follow-up studies in randomized patients (randomized clinical trial: MF diagnosed by CMR and no LV dysfunction) and nonrandomized patients (registry: without MF or with LV dysfunction).
There was a significant increase in MF mass over time in group D (n = 11) but not in group ND-NF (n = 21) (Figure 2A and Table 2). We observed a significant positive correlation between age and the amount of MF identified by CMR both at baseline and at follow-up (correlation coefficients, r = 0.52 at baseline and r = 0.50 at follow-up; P < .001 for both). We also observed a significant negative correlation between age and CK levels (r = −0.693; P < .001). In addition, there was an inverse association between CK level and MF at baseline (CK level decreased with age; MF identified by CMR increased with age) and, as expected, between LVEF and MF (as MF increased, LVEF decreased) (Figure 2B). Patients receiving corticosteroids showed no significant differences in baseline MF or MF progression compared with those not receiving corticosteroid therapy. The pattern of MF was mostly midwall and subepicardial and predominantly affecting lateral and inferior LV segments (Figure 3). Myocardial fibrosis measurements showed low intraobserver and interobserver variability and excellent agreement (mean difference, 0.06 g; limits of agreement, 2.4 to 2.5 g; coefficient of variation, 6.4%; and intraclass correlation coefficient, 0.99; vs mean difference, −0.56 g; limits of agreement, −5.1 to 4.0 g; coefficient of variation, 11.9%; and intraclass correlation coefficient, 0.98).
All 4 patients (5%) who died during follow-up had MF identified by CMR, and the mean amount of MF was significantly higher in these individuals than among the 72 survivors (95%) (36.4 [16.2] vs 11.0 [12.1] g; P < .001). In addition, LVEF was significantly lower on CMR among the patients who died (35.9% [16.3%] vs 55.5% [9.0%]; P < .001). All 4 deaths were related to heart failure associated with pneumonia, thromboembolic events, and severe cardiac arrhythmia. There were 8 patients with cardiovascular events, all of which related to cardiac arrhythmias (4 supraventricular and 4 ventricular tachycardia) requiring hospitalization to revert and/or control their arrhythmias either through electrical or chemical cardioversion. On the Kaplan-Meier analysis for the combined cardiovascular events, patients with MF showed a higher probability of experiencing an event compared with patients without MF (event rate, 10 of 55 [18.2%] vs 0 of 21 [0%]; log-rank P = .04) (Figure 2C).
Moreover, for all patients, the use of combined traditional diagnostic methods (chest radiograph, electrocardiogram, and echocardiogram) provided low sensitivity and negative predictive value for detecting cardiac involvement, defined as MF by CMR (sensitivity, 47.3%; specificity, 71.4%; positive predictive value, 81.2%; and negative predictive value, 34.1%).
We used a backward stepwise multivariate analysis to investigate indicators of MF progression by CMR and indicators of LVEF change on follow-up CMR study. For model 1 in Table 3, including age and MF (as percentage of LV mass) in the baseline CMR study, the indicators for the progression of MF were CK level and use of ACE inhibitors. Indicating factors for change in LVEF at follow-up CMR were MF in the baseline CMR and use of ACE inhibitors (Table 3).
Forty-two patients were included in the randomized clinical trial: 39 patients with DMD and 3 with BMD. Mean (SD) age in this cohort was 12.1 (2.7) years. All patients had their diagnosis confirmed by muscle biopsy results (25 [60%]) or DNA mutation (21 [50%]). Characteristics of this cohort are reported in Table 1.
At the 2-year follow-up, the increase in MF was significantly higher in the nontreated (ND-NT) compared with the treated (ND-T) group (10.0% [6.2%] vs 3.1% [7.4%]; P = .001) (Table 2). The increase in MF was not statistically significant in the ND-T group (32.8 [13.7] to 35.9 [13.7] percentage of LV mass; P = .07) but was significant in the ND-NT group (23.8 [13.6] to 33.8 [12.4] percentage of LV mass; P < .001) (Table 2 and Figure 2A). Patients receiving corticosteroids demonstrated no significant differences in baseline MF or MF progression compared with those not receiving corticosteroid therapy within randomized groups. Examples of MF patterns, distribution, and its progression and quantitative measurement are shown in Figure 3.
Model 2 for the backward stepwise multivariate analysis including only randomized patients had age and baseline MF (as percentage of LV mass) in the initial model for adjustment. Factors as indicators for progression of MF were CK value and treatment (ND-T) with a negative coefficient, that is, indicating less MF (coefficient [SE], −4.51 [2.11]; P = .04). The treated arm of randomization (ND-T) was the only indicator of preserved LVEF at follow-up CMR (P = .09) (Table 3).
This randomized clinical trial investigated the effect of intervention with ACE inhibitor treatment on the progression of MF quantified by CMR as the primary end point. To our knowledge, this is the first study to demonstrate the benefit of ACE inhibitor therapy specifically in patients with MF identified on CMR and preserved LV function. Patients with DMD or BMD with preserved LV function and MF demonstrated slower MF progression over a 2-year follow-up when receiving an ACE inhibitor. Moreover, patients with MF had a significantly higher probability of cardiovascular events than did those without MF in the 2-year follow-up (P = .04) (Figure 2C).
Since what we believe to be the first study describing MF detection by CMR in patients with DMD or BMD published by our group,7 other studies20,21 have confirmed a high prevalence of MF in these patients. In the present study, 55 patients (72%) presented with MF at baseline and, of those, only 13 individuals (24%) had abnormal LVEF shown on echocardiogram, highlighting the critical role of early detection of MF, particularly considering that the slower progression of MF over 2 years occurred only in patients with MF and no LV dysfunction at baseline. Currently, the therapeutic intervention with ACE inhibitors is initiated based on altered LVEF, which is a late finding in the myocardial damage cascade in muscular dystrophies. Our data suggest that ACE inhibitor therapy should start before the onset of LV dysfunction and when MF becomes detectable by CMR. This finding is further supported by an animal study by Rafael-Fortney et al22 showing that early intervention with antifibrotic drugs (eg, lisinopril, spironolactone) could attenuate myocardial damage in treated mice, as measured by circumferential strain and matrix metalloproteinase activity. Although the focus of our study was on patients with DMD or BMD who had preserved LVEF, for the entire patient group, including patients with reduced LVEF, the amount of MF showed a significant inverse correlation to LVEF (Figure 2B), which agrees with a recent study investigating the progression of LV dysfunction.23
Several studies tested the use of ACE inhibitors and other drugs in the prevention of LV dysfunction in patients with DMD or BMD,24-26 but, to our knowledge, no randomized clinical trial tested the benefit of adding these medications based on MF detection by CMR in a 2-year follow-up. In the multivariate analysis, the use of ACE inhibitors was an independent indicator of less MF at the follow-up (P = .04) and the only indicator of improved LVEF at follow-up but with a nonsignificant P value (P = .09) (Table 3).
Given the rare occurrence of DMD and BMD, this trial had a small number of randomized patients. Prognostic data based on few events were limited. Sample size calculation was also limited by few data on progression of MF absolutely quantified in grams, although a large data set published by Tandon et al23 looked at MF progression as the number of LV segments. Nonetheless, our outcome data can help with hypothesis generation and sample size calculation for future randomized studies.
Even though patients were randomized to receive or not receive ACE inhibitor therapy, some variables were not completely equalized because of the relatively small number of patients, particularly the ND-NT subgroup, which was younger (mean [SD] age, 11.1 [2.3] years) than the ND-T subgroup (13.1 [2.7] years). This age difference might, at least in part, account for the observed difference in MF progression between both subgroups over the 2-year follow-up; however, there is evidence suggesting that this was not the case. In the other nonrandomized patient subgroups (D and ND-NF), the older D subgroup (19.2 [3.9] years) demonstrated a high rate of MF progression, and the younger ND-NF subgroup (11.2 [4.2] years) experienced the slowest rate of MF progression. Multivariable analyses demonstrated that, even after correcting for age, randomization status remained a significant predictor of MF progression. Therefore, we believe that the small age imbalance did not account for the difference in MF progression observed between the ND-NT and ND-T subgroups.
In patients with DMD and BMD with preserved LV function and MF diagnosed by CMR, the use of ACE inhibitors slows MF progression at a 2-year follow-up. The presence of MF indicates a higher probability of cardiovascular events in a 2-year follow-up.
Corresponding Author: Carlos Eduardo Rochitte, MD, PhD, Heart Institute, InCor, University of São Paulo Medical School, Avenida Dr. Enéas de Carvalho Aguiar, 44, Andar AB, Cerqueira César, São Paulo, SP, Brazil, 05403-000 (email@example.com).
Accepted for Publication: October 13, 2016.
Published Online: December 7, 2016. doi:10.1001/jamacardio.2016.4801
Author Contributions: Drs Silva and Rochitte had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Silva, Magalhães, Meira, Azevedo, Gurgel-Giannetti, Kalil-Filho, Rochitte.
Acquisition, analysis, or interpretation of data: Silva, Meira, Rassi, Andrade, Gutierrez, Vainzof, Zatz, Rochitte.
Drafting of the manuscript: Silva, Magalhães, Gutierrez, Azevedo, Rochitte.
Critical revision of the manuscript for important intellectual content: Silva, Magalhães, Meira, Rassi, Andrade, Azevedo, Gurgel-Giannetti, Vainzof, Zatz, Kalil-Filho, Rochitte.
Statistical analysis: Silva, Andrade, Azevedo, Rochitte.
Obtained funding: Rochitte.
Administrative, technical, or material support: Silva, Magalhães, Meira, Rassi, Gutierrez, Gurgel-Giannetti, Kalil-Filho, Rochitte.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Funding/Support: This study was funded in part by Zerbini Foundation (Dr Rochitte) and received funding from São Paulo Research Foundation–Centers of Research, Innovation and Diffusion and National Council for Scientific and Technological Development–National Institutes of Science and Technology for the genetic evaluation (Drs Vainzof and Zatz).
Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.