Author + information
- Received April 9, 2018
- Revision received June 28, 2018
- Accepted July 29, 2018
- Published online September 24, 2018.
- Gisela Teixido-Tura, MD, PhDa@giselateixido,
- Alberto Forteza, MD, PhDb,
- Jose Rodríguez-Palomares, MD, PhDa@JRodriPalomares,
- Jesús González Mirelis, MD, PhDb,c,d,
- Laura Gutiérrez, MDa,
- Violeta Sánchez, MD, PhDe,
- Borja Ibáñez, MD, PhDc,d,f,
- David García-Dorado, MD, PhDa and
- Artur Evangelista, MD, PhDa,∗ (, )@cardioimagenVH
- aServei de Cardiologia, Hospital Universitari Vall d’Hebron, CIBERCV, Barcelona, Spain
- bHospital Puerta de Hierro, Majadahonda, Spain
- cCentro Nacional de Investigaciones Cardiovasculares Carlos III (CNIC), Madrid, Spain
- dCentro de Investigación Biomédica en Red Enfermedades Cardiovaculares, Madrid, Spain
- eHospital Universitario 12 de Octubre, Centro de Investigación Biomédica en Red Enfermedades Cardiovaculares, Madrid, Spain
- fIIS-Fundación Jiménez Díaz, Madrid, Spain
- ↵∗Address for correspondence:
Dr. Artur Evangelista, Department of Cardiology, Hospital Universitari Vall d’Hebron, Passeig Vall d’Hebron 119-129, 08035, Barcelona, Spain.
Background Beta-blockers are the standard treatment in Marfan syndrome (MFS). Recent clinical trials with limited follow-up yielded conflicting results on losartan’s effectiveness in MFS.
Objectives The present study aimed to evaluate the benefit of losartan compared with atenolol for the prevention of aortic dilation and complications in Marfan patients over a longer observation period (>5 years).
Methods A total of 128 patients included in the previous LOAT (LOsartan vs ATenolol) clinical trial (64 in the atenolol and 64 in the losartan group) were followed up for an open-label extension of the study, with the initial treatment maintained.
Results Mean clinical follow-up was 6.7 ± 1.5 years. A total of 9 events (14.1%) occurred in the losartan group and 12 (18.8%) in the atenolol group. Survival analysis showed no differences in the combined endpoint of need for aortic surgery, aortic dissection, or death (p = 0.462). Aortic root diameter increased with no differences between groups: 0.4 mm/year (95% confidence interval: 0.2 to 0.5) in the losartan and 0.4 mm/year (95% confidence interval: 0.3 to 0.6) in the atenolol group. In the subgroup analyses, no significant differences were observed considering age, baseline aortic root diameter, or type of dominant negative versus haploinsufficient FBN1 mutation.
Conclusions Long-term outcome of Marfan syndrome patients randomly assigned to losartan or atenolol showed no differences in aortic dilation rate or presence of clinical events between treatment groups. Therefore, losartan might be a useful, low-risk alternative to beta-blockers in the long-term management of these patients.
Marfan syndrome (MFS) is a hereditary connective tissue disorder caused by mutations in the FBN1 gene, which encodes for fibrillin-1 protein, an important constituent of the extracellular matrix. Although these patients have a multisystemic disease that involves ocular, skeletal, and cardiovascular systems, survival is mainly determined by aortic complications. The current management of aortic involvement in MFS includes regular aortic imaging to evaluate aortic dilation progression, and prophylactic aortic repair when aortic dilation reaches a defined threshold that carries the threat of dissection (1). Current medical treatment aims to delay aortic dilation progression and avoid aortic complications, and beta-blockers are the standard treatment for preventing aortic dilation. However, the randomized evidence supporting the use of beta-blockers is limited to a single, small, open-label trial (2), and many patients experience progressive aortic dilation despite such therapy (3,4). Furthermore, aortic dilation in MFS has been related to overexpression of transforming growth factor–beta (5). Therefore, losartan, an angiotensin-II receptor blocker (ARB) that has previously demonstrated transforming growth factor–beta antagonism, has been studied. Four large clinical trials with different designs, either comparing losartan with beta-blockers (6,7) or adding losartan or placebo to baseline treatment (8,9), included Marfan patients with different age range and aortic dilation profiles. Results at 3 years of follow-up ranged from reduction of aortic dilation from losartan treatment to no benefit of losartan added to baseline treatment (most of them with baseline beta-blockers). In the LOAT (LOsartan vs ATenolol) trial (7), similar to The Pediatric Heart Network trial (6), no differences in aortic diameter progression were observed with losartan compared with atenolol treatment. The conflicting results of the published studies raised the question of whether longer intervention would yield different results. Therefore, the aim of the present study was to evaluate the benefit of losartan compared with atenolol for the prevention of aortic dilation and aortic complications in Marfan patients over a longer period of observation (>5 years).
The MFS patients included in the previous clinical trial (7) were followed-up over an open-label extension of the study, maintaining the initial treatment assigned by randomization. The follow-up protocol included a minimum of an annual clinical and echocardiographic evaluation. Clinical events (need for aortic surgery based on Guidelines recommendations , aortic dissection, and/or death) were registered and, after at least 5 years had elapsed from the start of the clinical trial, a cardiac magnetic resonance imaging (CMR) study was performed. The study was approved by the ethics committees of both hospitals, and all patients (or their guardians) provided their written informed consent.
The initial inclusion criteria for participation in the LOAT trial were: MFS diagnosis (based on original Ghent criteria), age between 5 and 60 years, and maximum aortic root diameter <45 mm. Exclusion criteria were: history of aortic dissection or cardiac/aortic surgery; grade III/IV aortic regurgitation; history of angioedema or any other intolerance to angiotensin-converting enzyme inhibitors, ARB, or beta-blockers; treatment with other cardiovascular drugs; pregnancy or planned pregnancy; or history of asthma, respiratory failure, kidney, or neurological disease. A total of 140 participants were included in the LOAT trial. In the present study, 12 patients were ruled out: 5 who withdrew consent, 1 for hypertension, 2 for planned pregnancy, and 2 for noncompliance within 6 months of the start of the trial (Figure 1), and 2 in whom Marfan diagnosis was posteriorly ruled out. Thus, 128 patients were included for the extension of the trial (n = 64 losartan group, n = 64 atenolol group).
Patients were maintained on the same drug (losartan or atenolol) and at the same dosage as during the clinical trial. Treatment was interrupted if pregnancy was planned (for losartan) or signs of clinical intolerance (hypotension or asthenia) or other side effects associated with the treatment appeared.
The CMR study was performed with the same protocol as in the LOAT trial. All CMR studies were conducted with 1.5-T (Signa Excite GE, Milwaukee, Wisconsin) and 3-T systems (Philips Achieva Tx, Best, the Netherlands). Imaging of the entire aorta was obtained: steady-state free precession cine acquisitions were obtained in held expiration and with electrocardiography synchronization. Perpendicular cine images were obtained at the level of the sinuses of Valsalva and ascending aorta (at the level of the pulmonary bifurcation) using the double-oblique technique. The 3 cusp-to-commissure diameters were measured at the aortic root level at the end-diastolic frame. The maximum of the 3 diameters at baseline and at follow-up was considered for analysis. Maximum ascending aortic diameter was also measured. All measurements were indexed by body surface area at the time of the CMR study. Measurements were taken by a blinded observer (N.V.) using the QFlow software package, version 5.5 (Medis, Leiden, the Netherlands).
As in the previous LOAT trial, primary endpoints were defined as changes in aortic root and ascending aorta diameter indexed by body surface area on CMR with losartan versus atenolol. Secondary endpoints were clinical events during follow-up, including aortic dissection, need for aortic intervention or heart surgery, and death.
Analyses were made following the modified intention-to-treat principle. Continuous variables were summarized as mean ± SD or median and interquartile range. Categorical variables were presented as counts and percentages. Comparison of continuous variables was made by Student’s t-test and by chi-square test for categorical variables. When normality was not assumed, nonparametric analyses were made instead. Survival curves were estimated using the Kaplan-Meier method and compared with log-rank tests. Statistical significance was considered at p ≤0.05 (2-sided). Analyses were performed using SPSS 19.0 software (IBM, Chicago, Illinois).
A total of 128 patients were included in the study: 64 from the losartan group and 64 from the atenolol group. Baseline characteristics are described in Table 1. No differences were observed related to age, sex, body surface area, or aortic dimensions. A total of 109 (85%) patients had FBN1 gene analysis, of whom a pathogenic mutation was found in 89 (82%): 50 (56%) were classified as dominant negative, 26 (29%) as haploinsufficient, and 13 (15%) as unclassifiable. In the losartan group, 54 patients (84.4%) continued with losartan throughout follow-up; of the rest, 4 discontinued medical treatment, 5 changed to beta-blockers, and 1 received a combination of atenolol plus losartan (due to hypertension). A total of 48 patients (75.0%) in the atenolol group continued with atenolol throughout follow-up; 12 of the remaining patients discontinued medical treatment, and 4 changed to losartan.
Mean clinical follow-up was 6.7 ± 1.5 years (median 7.2 years; interquartile range: 6.3 to 7.8 years). Nine events occurred in the losartan group (7 elective ascending aorta surgery, 1 elective descending aorta surgery, and 1 type A aortic dissection) and 12 in the atenolol group (9 elective ascending aorta surgery, 2 type A aortic dissection, and 1 type B aortic dissection). Aortic root diameters prior to type A dissections were 43, 47, and 49 mm. Of the 4 aortic dissections, 3 resulted in death (1 type B). Survival analysis showed no differences in the combined endpoint of aortic surgery, aortic dissection, or death (p = 0.462) or for the combined endpoint of acute aortic syndrome or death (p = 0.305). Survival curves are shown in the Central Illustration.
A total of 61 patients from the losartan group and 58 from the atenolol group had follow-up CMR, either at a minimum of 5 years of follow-up or before the event. Mean CMR follow-up was 5.9 ± 2.0 years. Aortic root diameter increased in both groups: 0.4 mm/year (95% confidence interval [CI]: 0.2 to 0.5 mm/year) in the losartan and 0.4 mm/year (95% CI: 0.3 to 0.6 mm/year) in the atenolol groups. However, no statistically significant differences in aortic dilation progression were observed between groups: −0.0 mm/year (95% CI: −0.25 to 0.17 mm/year; p = 0.754). Furthermore, no statistically significant differences were observed when evaluating aortic root diameters indexed by body surface area or ascending aorta diameters (Table 2). Similar results were found when only patients who continued with the same treatment as during the trial were considered (Online Table 1).
In the subgroup analyses, no significant beneficial effect of losartan was observed on aortic root diameter regardless of sex, age group, baseline aortic root diameter, or type of dominant negative versus haploinsufficient FBN1 mutation (Figure 2).
This is the longest study to date comparing clinical evolution and aortic dilation in Marfan patients treated with atenolol versus losartan. This study found no differences in clinical events over a long period of treatment (median: 7.2 years) or in aortic dilation progression assessed by CMR (median 6.7 years). Furthermore, no differences were observed in the subgroup analysis by sex, age, baseline aortic diameters, or type of FBN1 mutation.
Beta-blockers have become standard in the prevention of aortic dilation and dissection in patients with MFS since the report by Shores et al. (2). However, their benefit is debatable, and 2 meta-analyses also reached opposing conclusions (10,11). More recently, 4 prospective randomized trials, COMPARE (COzaar in Marfan PAtients Reduces aortic Enlargement) (8), Pediatric Heart Network (6), SARTAN (Marfan Sartan) (9), and LOAT (7), were the first to examine the use of ARBs in Marfan patients; these trials followed the publication of the landmark work by Habashi et al. (12), in a mouse model, and the remarkably positive results of a small, nonrandomized trial on the effects of losartan in a small subset of children with a severe phenotype of the disease (3). The designs of these recent trials were diverse: the COMPARE and SARTAN trials assessed the benefit of adding losartan to baseline therapy compared with adding placebo, and the Pediatric Heart Network and LOAT trials compared the administration of losartan versus atenolol. The COMPARE was open-label, Pediatric Heart Network was single-blinded, and SARTAN and LOAT were double-blinded, randomized trials. Evaluation of the primary outcome, the rate of aortic root enlargement, was evaluated by echocardiography in the largest trials, the Pediatric Heart Network and SARTAN, and by CMR in the COMPARE and LOAT trials. Although COMPARE included only adult patients, 28% and 40% of patients in the SARTAN and LOAT trials, respectively, were <18 years of age. By contrast, 75% of patients in the Pediatric Heart Network were <16 years, and the mean Z-score of included patients ranged from 3.2 to 4.0. Clinical events were similar between treatment groups, and only the COMPARE trial showed a slight reduction in aortic root growth rate at 3 years in the group treated with losartan (0.77 ± 1.36 mm vs. 1.35 ± 1.55 mm; p < 0.014). No relationship was found between blood pressure changes during treatment and aortic root growth rate changes. Recently, a meta-analysis was published of all prospective randomized clinical trials studying the effect of losartan on aortic dilation (13). The results indicate that losartan treatment slows the rate of aortic dilation; however, no benefits on clinical outcomes were observed in the losartan group compared with the nonlosartan group (most under beta-blocker treatment). This result, however, should be interpreted with caution: as previously exposed, the design of the 6 studies differed significantly, and the meta-analysis included redundant information of subgroup analyses of the COMPARE trial. Furthermore, the relatively low dose of losartan administered in human trials compared with the mouse model could be an explanation for the lack of consistent positive results in human. However, one of the questions that remains to be answered is whether a longer intervention would yield different results. This question is partly addressed in the current study, where most patients maintained the baseline randomized treatment and its doses, and after a 6-year treatment, we found no difference between losartan and atenolol treatment groups in the aortic root dilation rate or clinical events. In addition, we found no differences in losartan efficacy depending on the type of FBN1 mutation, thus not replicating the results from Franken et al. (14), which demonstrated a more marked reduction of aortic dilation rate with the addition of losartan in haploinsufficient Marfan patients.
Although the initial trial was double-blinded, the extension was open-label. However, experts who measured aorta size by CMR were blinded to the medical treatment and clinical information. A further limitation was the absence of a control group with placebo, which was not possible for ethical reasons because beta-blockers are accepted as the standard treatment. The final sample size of the study was not large; however, with the use of CMR for aortic diameter measurements, it sufficed to detect clinically significant differences of 2 mm in 6 years in the aortic root dilation rate between treatment groups (97% statistical power to detect a difference ≥0.4 mm/year). In the current study, 85% of patients underwent FBN1 gene analysis, 82% of whom had pathogenic mutations. The proportion of negative FBN1 in our study (18%) was similar to other clinical trials: 15% in the SARTAN, 16.2% in the COMPARE, and 11.7% in the Pediatric Heart Network.
The results showed no differences in long-term treatment with losartan compared to atenolol in aortic root and ascending aorta dilation rate or the frequency of clinical events in MFS patients. Although losartan was not proven superior to atenolol in monotherapy in MFS patients, it might be a useful, low-risk alternative to beta-blockers in the long-term management of these patients.
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: In patients with MFS, the ARB losartan is a reasonable alternative to beta-adrenergic inhibition with atenolol for prevention of aortic dilation and associated complications.
TRANSLATIONAL OUTLOOK: Further studies are needed to evaluate the utility of combined treatment with both drugs in this situation.
The authors are indebted to Christine O’Hara for help with the English version of the manuscript; to our nurses, Francesca Huguet (Servei de Cardiologia, Hospital Universitari, Vall d’Hebron, Barcelona), Pilar Carrascal, and Sonia Casado (Marfan Unit, Hospital Universitario 12 de octubre, Madrid), for their collaboration in the study; and especially to all of the patients who have participated in the study. Part of the magnetic resonance imaging studies were performed at the TRIMA@CNIC (Translational IMAging at Centro Nacional de Investigaciones Cardiovasculares) node of the ICTS (Infraestructuras Científicas y Técnicas Singulares) REDiB.
This work was been funded by a grant of the Spanish Society of Cardiology and CIBERCV. The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- angiotensin-II receptor blocker
- confidence interval
- cardiac magnetic resonance imaging
- Marfan syndrome
- Received April 9, 2018.
- Revision received June 28, 2018.
- Accepted July 29, 2018.
- 2018 American College of Cardiology Foundation
- Erbel R.,
- Aboyans V.,
- Boileau C.,
- et al.,
- for the Task Force for the Diagnosis and Treatment of Aortic Diseases of the European Society of Cardiology (ESC)
- Matt P.,
- Schoenhoff F.,
- Habashi J.,
- et al.
- Habashi J.P.,
- Judge D.P.,
- Holm T.M.,
- et al.
- Gao L.,
- Chen L.,
- Fan L.,
- et al.
- Franken R.,
- den Hartog A.W.,
- Radonic T.,
- et al.