Clinical Stratification of Pediatric Patients with Idiopathic Thoracic Aortic Aneurysm Benjamin J. Landis, MD1, Stephanie M. Ware, MD, PhD2, Jeanne James, MD1, Amy R. Shikany, MS1, Lisa J. Martin, PhD3, and Robert B. Hinton, MD1 Objectives To describe the global phenotypes of pediatric patients with thoracic aortic aneurysm (TAA) who do not have a clinical diagnosis of Marfan syndrome (MFS) or related connective tissue disorders. We hypothesized that the presence of noncardiovascular abnormalities correlate with TAA severity and that medical therapy reduces TAA progression. Study design This is a retrospective case series of patients with TAA age #21 years evaluated in a cardiovascular genetics clinic. Patients meeting clinical criteria for MFS or related disorders were excluded. Repeated measures analyses of longitudinal echocardiographic measurements of the aorta were used to test associations between TAA severity and noncardiovascular phenotype and to assess the impact of medical therapy. Results Sixty-nine patients with TAA at mean age 12.5 5.3 years were included. Noncardiovascular abnormalities, including skeletal (65%) or craniofacial (54%) findings, were frequently observed. Increased rate of aortic root enlargement was associated with ocular (P = .002) and cutaneous (P = .003) abnormalities, and increased rate of ascending aorta enlargement was associated with craniofacial (P < .001) abnormalities. Beta blocker or angiotensin receptor blocker therapy (n = 41) was associated with reduction in the rate of aortic root growth (P = .018). Conclusions Children with TAA not satisfying diagnostic criteria for MFS or related disorders frequently have noncardiovascular findings, some of which are associated with TAA progression. Because therapy initiation may reduce risk of progression and long-term complications, comprehensive assessment of noncardiovascular findings may facilitate early risk stratification and improve outcomes. (J Pediatr 2015;167:131-37). See editorial, p 14
T
horacic aortic aneurysm (TAA) is a subclinical aortopathy characterized by dilation of the proximal aorta. TAA is associated with severe acute complications such as aortic dissection and sudden cardiac death. There are an estimated 30 000 deaths per year in the US because of aneurysmal disease.1 Pediatric TAA is well recognized in connective tissue disorders (CTDs) such as Marfan syndrome (MFS) and Loeys-Dietz syndrome (LDS), but pediatric patients with TAA who do not meet diagnostic criteria for these conditions have not been well characterized. There is, however, mounting evidence to indicate that the aortopathy in these patients also has a genetic and developmental basis. The 2010 guidelines from the American Heart Association provide thresholds for prophylactic replacement of the aorta, based primarily on aortic diameter or rapid aortic enlargement.2 In these guidelines, the threshold for surgery is lower for patients with LDS than MFS and is lower for patients with MFS than “nonsyndromic” TAA, but the stratification of patients who do not have a diagnosis of a specific CTD is limited. In the context of increasing recognition of the genetic basis of TAA, there is a need to characterize this heterogeneous patient population with idiopathic TAA in order to develop risk classification paradigms to guide clinical management. Children with TAA may benefit from medical intervention to prevent disease progression, cardiac events, and need for surgery. We have performed deep phenotyping of cardiovascular and noncardiovascular characteristics in a pediatric cohort of patients with TAA who do not have the diagnosis of a specific CTD such as MFS or LDS. We aimed to generate a phenotype-based framework for risk classification and hypothesized that there are noncardiovascular abnormalities associated with TAA severity.
ARB BAV BB CTD CVM FTAAD LDS MFS MPA STJ TAA
Angiotensin II type 1 receptor blocker Bicuspid aortic valve Beta blocker Connective tissue disorder Cardiovascular malformation Familial thoracic aortic aneurysm or dissection Loeys-Dietz syndrome Marfan syndrome Main pulmonary artery Sinotubular junction Thoracic aortic aneurysm
From the 1Division of Cardiology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH; 2Department of Pediatrics and Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, IN; and 3 Division Human Genetics, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH Supported by the American Heart Association (IRG 18830027 [to R.H.]) and the Cincinnati Children’s Hospital Medical Center’s Research Foundation (to R.H.). The authors declare no conflicts of interest. 0022-3476/$ - see front matter. Copyright ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2015.02.042
131
THE JOURNAL OF PEDIATRICS
www.jpeds.com
We additionally hypothesized that beta blocker (BB) or angiotensin II type 1 receptor blocker (ARB) therapy reduces the rate of aortic dilation. Consideration of the patient’s global phenotype may be useful to stratify risk and guide clinical interventions.
Methods This was a retrospective cohort study consisting of review of existing medical records; the local Institutional Review Board approved this study. Patients with TAA age 21 years or younger were identified by review of cardiovascular genetics clinic rosters from July 1, 2010, to December 31, 2012. TAA was defined as aortic root or ascending aorta z-score greater than +2 by transthoracic echocardiography.3 All patients underwent evaluation by a pediatric cardiologist, medical geneticist, and genetic counselor, which included comprehensive physical examination and detailed pedigree ascertainment. The patients evaluated in this clinic constitute a phenotypically and genetically heterogeneous spectrum including syndromic TAA (eg, MFS, LDS, Turner syndrome), familial TAA or dissection (FTAAD), patients with some concern for a CTD, and patients referred for coordinated management of isolated TAA. Patients with syndromic TAA were excluded. The included patients were further classified by a medical geneticist (S.W.) utilizing the systemic score of the revised Ghent nosology for MFS as having no or low (score 0-1), intermediate (score 2-4), or high (score 5-6) suspicion for MFS.4 In addition to excluding patients with the diagnosis of a specific CTD or syndrome associated with TAA, patients with complex cardiovascular malformations (CVMs) associated with aortic dilation (eg, tetralogy of Fallot) or at least moderate aortic valve stenosis or regurgitation were also excluded. Data collected from the electronic medical record included demographics, anthropomorphic data, vital signs, past medical history, family history, physical examination findings including Beighton score (hypermobility), noncardiovascular features of LDS, and systemic score of the revised Ghent nosology (MFS),4,5 medication use, and genetic testing. At each visit the patient was examined for the development of new findings, which were included in these analyses. Echocardiograms performed from May 1, 2003, to December 31, 2012 were reviewed. Data extracted included aorta and main pulmonary artery (MPA) diameters, aortic and mitral valve morphology and function, and presence of additional CVMs. To improve reliability, aortic dimensions were measured in triplicate by a single reader (B.L.) and 10% were re-measured by a second reader (R.H.). Aortic measurements were obtained with 2-dimensional echocardiography in the parasternal long axis in systole, according to consensus guidelines.2,6 Aortic annulus, aortic root, sinotubular junction (STJ), proximal ascending aorta, and MPA diameters were measured from inner edge to inner edge perpendicular to the long axis of the vessel at maximal expansion during systole. Absolute measurements were indexed to 132
Vol. 167, No. 1 body surface area, using the Haycock formula,7 and normalized to generate z-scores based on reference data from Boston Children’s Hospital for annulus and root8,9 and local data for STJ and ascending aorta. Intervals of therapy and doses were recorded. Atenolol was the prescribed BB in all but 1 patient (metoprolol), and losartan was the only ARB prescribed. Patients were divided into 2 groups: no therapy or therapy (BB or ARB). For exploratory subanalysis, the therapy group was subdivided into BB only or ARB only. Patients with reported noncompliance were considered as not receiving therapy. Statistical Analyses Analyses were performed using JMP statistical software package (SAS Institute, Cary, North Carolina). Continuous data were evaluated for distributional qualities. We examined the variability in the triplicate measures of echocardiographic data using coefficient of variation (cardiovascular). When triplicate values exceeded 5% cardiovascular, the 2 most similar measures were selected. Once consistency was evaluated, the maximum value of the retained replicated measure was used. In addition, 10% of studies were independently remeasured by a second reader (R.H.) with strong correlation (R2 > 0.95) between readers. Baseline characteristics of the cohort were described. For continuous variables distributed normally (eg, age), means and SDs are reported. For continuous variables that were non-normally distributed (eg, echocardiogram intervals, revised Ghent systemic score), medians, and IQRs are used. Categorical data are reported as frequencies. To identify differences between treatment groups, we utilized t tests for continuous normally distributed variables, Wilcoxon rank sum for continuous non-normally distributed variables, and contingency tables and c2 goodness of fit tests for categorical variables. To examine longitudinal changes in aortic dimensions from baseline study, repeated measures analysis (mixed models) was used on z-scores. For individuals in the no therapy group, the baseline study was considered the first echocardiogram. For individuals receiving BB or ARB, the baseline study was considered the last echocardiogram prior to therapy initiation. The primary outcome variables were zscores of the aortic root and ascending aorta. Time was the primary fixed effect with the individual as a random effect. Other covariates included age and presence of noncardiovascular or valvar abnormalities. As we considered 11 distinct noncardiovascular or valvar abnormalities, we employed a Bonferroni correction (a = 0.05/11 = 0.0045) for statistical significance of the rate of aortic root change for noncardiovascular and valvar abnormalities. To test whether there were differences in disease progression between untreated and treated individuals prior to the initiation of treatment, an interaction of treatment status and time was included with the analysis restricted to untreated and pretreatment treated individuals. To test whether initiation of treatment changed progression, we performed an analysis in treated individuals with an interaction term between pre-post Landis et al
ORIGINAL ARTICLES
July 2015 treatment and time. No multiple testing correction was applied for evaluating effect of medical therapy.
Results The medical history of each of the 128 pediatric patients with TAA who were evaluated in cardiovascular genetics clinic over the study period was reviewed. After excluding those diagnosed with a genetic syndrome associated with TAA (n = 48), complex CVM (n = 7), or worse than mild aortic valve stenosis or regurgitation (n = 4), there were 69 (53%) patients who comprised the study population (Figure 1). Referral indications for the study population were evaluation and management of suspected CTD or other genetic disorder in 45 (65%), evaluation and management of TAA without referring caregiver having clear concern for CTD in 18 (26%), and family history of TAA or dissection in 6 (9%) patients. There was a median of 2 (range 1-5) clinic visits and median of 4 (range 1-18) echocardiograms per subject. The study population consisted predominantly of white males in late childhood to adolescence (Table I). The population was slender with mean body mass index of 18 kg/m2 and included 15 (22%) patients exceeding 95th percentile for height. The median systemic score according to the revised Ghent nosology was 2 (IQR 1-4), which is below the diagnostic threshold of 7. Based upon individual systemic scores, there were 30 (43%) patients with low or no suspicion for MFS (score 0 or 1), 31 (45%) with intermediate suspicion for MFS (score 2-4), and 8 (12%) with high suspicion for MFS (score 5 or 6) (Figure 1 and Table II; Table II is available at www.jpeds.com). Although MFS was highly suspected in only a minority of the cohort,
Table I. Patient characteristics, cardiovascular, and noncardiovascular findings according to therapy group Patient characteristics
n (%) No therapy BB or ARB (N = 69) (N = 28) (N = 41)
Demographics Age at initial visit (y) 12.5 (5.3) Sex Male 60 (87) Female 9 (13) Race White 60 (87) Black 6 (9) Asian 1 (1) Other 2 (3) Ethnicity* Non-Hispanic 66 (97) Hispanic 3 (3) Anthropomorphic data, mean (SD) Height percentile 58 (33) Weight percentile 45 (30) BMI absolute 18 (2.6) BMI percentile 35 (26) BSA absolute 1.4 (0.5) Blood pressure, mean (SD) Systolic percentile 56 (27) Diastolic percentile 60 (23) Noncardiovascular abnormalities Skeletal 45 (65) Craniofacial 37 (54) Ocular 22 (32) Neurodevelopmental 17 (25) Cutaneous 13 (19) Gastrointestinal 7 (10) Pulmonary 3 (3) Cardiovascular abnormalities Aortic root z-score +2 to 3 22 (32) z-score +3 to 4 21 (30) z-score > +4 21 (30) Ascending aorta† z-score +2 to 3 18 (29) z-score +3 to 4 7 (11) z-score > +4 12 (19) BAV 17 (25) MVP 11 (16)
P
11.9 (5.9)
12.9 (4.9)
.51
21 (75) 7 (25)
39 (95) 2 (5)
.03
25 (89) 2 (7) 0 1 (4)
35 (85) 4 (10) 1 (2) 1 (2)
.73
27 (100) 0
39 (95) 2 (5)
.51
52 (35) 40 (32) 18 (2.5) 32 (26) 1.3 (0.5)
61 (32) 49 (29) 18 (2.6) 37 (27) 1.4 (0.4)
.27 .23 .74 .48 .48
55 (23) 61 (21)
57 (29) 60 (25)
.7 .9
16 (57) 16 (57) 9 (32) 7 (25) 3 (11) 2 (7) 2 (7)
29 (71) 21 (51) 13 (32) 10 (24) 10 (24) 5 (12) 1 (2)
.25 .63 .97 .95 .21 .69 .56
12 (43) 6 (21) 5 (18)
10 (24) 15 (37) 16 (39)
.18
6 (23) 1 (4) 4 (15) 7 (25) 4 (14)
12 (32) 6 (16) 8 (22) 10 (24) 7 (17)
.71 .95 .76
BMI, body mass index; BSA, body surface area; MVP, mitral valve prolapse. P represents comparison of no therapy vs BB or ARB. *One ethnicity not known. †Unable to measure ascending aorta dimension for 6 patients.
Figure 1. Distribution of patients with TAA in cardiovascular genetics clinic population. The clinic patient population with TAA (N = 128) was comprised of patients with syndromic TAA, CVM associated with TAA, moderate to severe aortic valve dysfunction, and the study cohort (n = 69). The study cohort (white) is subdivided with dashed lines into categories of high (H), intermediate (I), and low/no (L) suspicion for MFS AV, aortic valve.
noncardiovascular abnormalities were frequently observed even among patients who were not referred for a suspected CTD. Altogether, more than 90% of patients were observed to have at least 1 noncardiovascular abnormality, including 45 (65%) patients with skeletal abnormalities, 37 (54%) with craniofacial abnormalities, and 22 (32%) with ocular abnormalities (Tables I and III; Table III is available at www.jpeds.com). Joint hypermobility, defined as Beighton score of 5 or greater, was infrequently observed (n = 7, 10%); none met clinical criteria for hypermobility EhlersDanlos syndrome. Among the study cohort, FBN1, TGFBR1, and TGFBR2 sequencing was negative for pathogenic variants in 42 (61%) patients, and additional deletion/duplication testing was negative in 3 patients (Table II). The majority (77%)
Clinical Stratification of Pediatric Patients with Idiopathic Thoracic Aortic Aneurysm
133
THE JOURNAL OF PEDIATRICS
www.jpeds.com
of patients in the study cohort who were determined to have intermediate or high suspicion for MFS had negative testing for FBN1, TGFBR1, and TGFBR2 mutations. At the time of last evaluation, 8 (12%) patients had a diagnosis of FTAAD based on confirmed family history of autosomal dominant TAA. Sequencing of ACTA2 and MYH11 was normal in 15 (22%) of the entire cohort. None of those with high suspicion for MFS or LDS was tested for ACTA2 and MYH11 mutations compared with 25% in the intermediate or low/no suspicion groups. These patterns indicate that over the study period, genetic testing was performed in a tiered manner based on suspicion for MFS or other CTD. Overall, the high frequency of noncardiovascular phenotype abnormalities in this collectively heterogeneous study cohort indicates that many pediatric patients with TAA have some degree of nonspecific systemic connective tissue involvement. The aortic root was dilated in 64 (93%) patients, but ascending aorta was dilated in 37 (59%) and annulus in only 3 patients (Table I). The mean z-score at baseline measurement was 3.1 1.1 (2.9 0.7 cm) for the aortic root and 2.0 1.7 (2.5 0.6 cm) for the ascending aorta (Figure 2). There were no cases of aortic dissection or surgical aortic replacement over the study period. Bicuspid aortic valve (BAV) was present in 17 (25%) patients, among whom 76% had fusion of the right and left coronary cusps, which is consistent with prior observations.10-12 Notably, 15 (88%) of the patients with BAV had aortic root dilation. In 11 of these cases of BAV, there was associated dilation of the ascending aorta; 2 patients had dilation restricted to the ascending aorta. Noncardiovascular abnormalities were also frequent in the BAV subset and similar to the overall cohort: 41% with skeletal, 53% with craniofacial, and 29% with ocular
Vol. 167, No. 1 abnormalities. The MPA was dilated in 19 (28%) patients, although image quality precluded measurement in a significant proportion of studies (43%), suggesting this may be an underestimate. All patients had normal left ventricular systolic function. Using longitudinal modeling (repeated measures), we observed significant associations between the rate of aortic enlargement and noncardiovascular abnormalities (Table IV). For example, ocular (P = .002) or cutaneous (P = .003) abnormalities were associated with an increased rate of aortic root dilation. Meanwhile, craniofacial (P < .001) abnormalities were associated with an increased rate of ascending aorta dilation. The severity of craniofacial dysmorphism, considered as the sum of craniofacial abnormalities in an individual, was also associated with rate of ascending aorta dilation (P < .001). There was no significant association between systemic score and rate of aortic enlargement or between BAV or mitral valve prolapse and the rate of aortic enlargement. When the cohort subset who had normal sequencing of FBN1, TGFBR1, and TGFBR2 were independently analyzed, the associations for cutaneous and craniofacial abnormalities with TAA progression remained significant, but ocular findings were no longer significant (Table V; available at www.jpeds.com). These observations provide a basis for stratifying the risk of TAA progression according to the pattern of a patient’s noncardiovascular abnormalities. BB or ARB therapy was initiated in 41 (59%) patients, including 24 (35%) on BB and 17 (25%) on ARB therapy. The mean age at therapy initiation was 12.9 4.9 years. The mean daily dose at the last follow up was 0.7 0.4 mg/kg for atenolol and 0.7 0.3 mg/kg for losartan (Table VI; available at www.jpeds.com). Although most patients were normotensive (Table I), 3 had systemic
Figure 2. A, Two-dimensional echocardiographic image of the proximal aorta. The aortic segments of interest are indicated by the double arrowhead lines and include: (1) aortic annulus; (2) aortic root; (3) STJ; and (4) ascending aorta. B, Bar plot demonstrating that the aortic root is more severely dilated at baseline compared with other aortic segments, and at baseline echocardiogram the patients who receive therapy have more severe aortic dilation than patients not receiving therapy. Bars represent mean z-score and error plots represent SDs from the mean. LA, left atrium; LV, left ventricle. 134
Landis et al
ORIGINAL ARTICLES
July 2015
Table IV. Noncardiovascular and cardiovascular factors impacting rate of aortic dilation
Affected systems Skeletal Craniofacial Craniofacial (sum) Ocular Neurodevelopmental Cutaneous Gastrointestinal Pulmonary Height >95th percentile Systemic score Cardiovascular BAV MVP
Relative change in rate of aortic root dilation 4 4 4 [[ 4 [[ 4 4 4 4 4 4
P
Relative change in rate of ascending aorta dilation
P
.4 .15 .54 .002 .88 .003 .10 .53 .37 .09
[ [[ [[ 4 4 Y 4 4 [ 4
.04 95th percentile Arm span/height >1.05 Low upper segment: lower segment (per Ghent criteria) Joint pain Pectus carinatum Decreased elbow extension Syndactyly Contractures Craniofacial Dolichocephaly Enophthalmos Downslanting palpebral fissures Malar hypoplasia Retrognathia Hypertelorism Palate abnormality Orthodontia Uvula abnormality Ear malformation Ocular Myopia Pupil/extraocular muscle abnormality Cutaneous Striae Easy bruising Abnormal scarring Gastrointestinal Inguinal hernia Eosinophilic esophagitis Pulmonary (pneumothorax) Cardiovascular abnormalities Aortic root z-score +2 to 3 z-score +3 to 4 z-score > +4
n (% of total)
No therapy (N = 28)
BB or ARB (N = 41)
BB (N = 24)
ARB (N = 17)
Pz
Px
12.5 (5.3)
11.9 (5.9)
12.9 (4.9)
13.1 (4.9)
12.5 (5.1)
.51 .026
.71 .17
60 (87) 9 (13)
21 (75) 7 (25)
39 (95) 2 (5)
24 (100) 0
15 (88) 2 (12) .73
.27
60 (87) 6 (9) 1 (1) 2 (3)
25 (89) 2 (7) 0 1 (4)
35 (85) 4 (10) 1 (2) 1 (2)
22 (92) 2 (8) 0 0
13 (76) 2 (12) 1 (6) 1 (6)
66 (97) 3 (3)
27 (100) 0
39 (95) 2 (5)
23 (96) 1 (4)
16 (94) 1 (6)
58 (33) 45 (30) 18 (2.6) 35 (26) 1.4 (0.5)
52 (35) 40 (32) 18 (2.5) 32 (26) 1.3 (0.5)
61 (32) 49 (29) 18 (2.6) 37 (27) 1.4 (0.4)
63 (32) 47 (30) 18 (2.2) 33 (27) 1.4 (0.4)
59 (33) 50 (28) 18 (3.1) 42 (27) 1.4 (0.5)
.27 .23 .74 .48 .48
.69 .76 .62 .27 .75
56 (27) 60 (23)
55 (23) 61 (21)
57 (29) 60 (25)
61 (33) 65 (23)
54 (25) 54 (27)
.7 .9
.49 .24
45 (65) 20 (29) 14 (20) 13 (19) 12 (17) 13 (19) 5 (7) 8 (12) 15 (22) 5 (7) 3 (4)
16 (57) 6 (21) 6 (21) 4 (14) 4 (14) 4 (14) 3 (11) 3 (11) 5 (18) 3 (11) 1 (4)
29 (71) 14 (34) 8 (20) 9 (22) 8 (20) 9 (22) 2 (5) 5 (12) 10 (24) 2 (5) 2 (5)
18 (75) 7 (29) 4 (17) 5 (21) 4 (17) 6 (25) 2 (8) 4 (17) 7 (29) 0 1 (4)
11 (65) 7 (41) 4 (24) 4 (24) 4 (24) 3 (17) 0 1 (6) 3 (17) 2 (12) 1 (6)
.25
.51
5 (7) 3 (4) 4 (6) 4 (6) 2 (3) 37 (54) 7 (10) 1 (1) 9 (13) 23 (33) 8 (12) 7 (10) 9 (13) 7 (10) 2 (3) 8 (12) 22 (32) 17 (25) 5 (7) 13 (19) 13 (19) 0 0 7 (10) 4 (6) 4 (6) 3 (4)
1 (4) 0 1 (4) 2 (7) 0 16 (57) 4 (14) 0 4 (14) 12 (43) 3 (11) 3 (11) 4 (14) 3 (11) 2 (7) 4 (14) 9 (32) 8 (29) 1 (4) 3 (11) 3 (11) 0 0 2 (7) 1 (4) 2 (7) 2 (7)
4 (10) 3 (7) 3 (7) 2 (5) 2 (5) 21 (51) 3 (7) 1 (2) 5 (12) 11 (27) 5 (12) 4 (10) 5 (12) 4 (10) 0 4 (10) 13 (32) 9 (22) 4 (10) 10 (24) 10 (24) 0 0 5 (12) 3 (7) 2 (5) 1 (2)
3 (13) 1 (4) 2 (8) 2 (8) 2 (8) 13 (54) 2 (8) 1 (4) 3 (13) 7 (29) 3 (13) 3 (13) 3 (13) 1 (4) 0 2 (8) 7 (29) 5 (21) 2 (8) 7 (29) 7 (29) 0 0 3 (12) 2 (8) 1 (4) 1 (4)
1 (6) 2 (12) 1 (6) 0 0 8 (47) 1 (6) 0 2 (12) 4 (24) 2 (12) 1 (6) 2 (12) 3 (18) 0 2 (12) 6 (35) 4 (24) 2 (12) 3 (18) 3 (18) 0 0 2 (12) 1 (6) 1 (6) 0
.63
.75
.97
.74
.21
.48
22 (32) 21 (30) 21 (30)
12 (43) 6 (21) 5 (18)
10 (24) 15 (37) 16 (39)
4 (17) 8 (33) 12 (50)
6 (35) 7 (41) 4 (24)
.51
Clinical Stratification of Pediatric Patients with Idiopathic Thoracic Aortic Aneurysm
1.0
.69
1.0
.56
1.0
.18 .11 (continued )
137.e4
THE JOURNAL OF PEDIATRICS
www.jpeds.com
Vol. 167, No. 1
Table III. Continued Patient characteristics
n (% of total)
No therapy (N = 28)
BB or ARB (N = 41)
BB (N = 24)
ARB (N = 17)
Pz
Px
.71 .95 .76
1.0 .79 .61
†
Ascending aorta z-score +2 to 3 z-score +3 to 4 z-score > +4 BAV MVP
18 (29) 7 (11) 12 (19) 17 (25) 11 (16)
6 (23) 1 (4) 4 (15) 7 (25) 4 (14)
12 (32) 6 (16) 8 (22) 10 (24) 7 (17)
7 (33) 4 (19) 5 (23) 5 (21) 3 (13)
5 (31) 2 (13) 3 (19) 5 (29) 4 (24)
BMI, body mass index; BSA, body surface area. *One ethnicity not known. †Unable to measure ascending aorta dimension for 6 patients. zNo therapy vs BB or ARB. xBB vs ARB.
Table VI. Therapy dosage of individual patients BB Age therapy started (y)
Table V. Non-cardiovascular and cardiovascular factors impacting rate of aortic dilation only among patients with negative testing for FBN1, TGFBR1, and TGFBR2
Affected systems Skeletal Craniofacial Craniofacial (sum) Ocular Neurodevelopmental Cutaneous Gastrointestinal Pulmonary Height >95th percentile Systemic score Cardiovascular BAV MVP
P
Relative change in rate of ascending aorta dilation
P
4 4 4 [ 4 [[ Y 4 4 4
.54 .96 .41 .03 .12
T
horacic aortic aneurysm (TAA) is a subclinical aortopathy characterized by dilation of the proximal aorta. TAA is associated with severe acute complications such as aortic dissection and sudden cardiac death. There are an estimated 30 000 deaths per year in the US because of aneurysmal disease.1 Pediatric TAA is well recognized in connective tissue disorders (CTDs) such as Marfan syndrome (MFS) and Loeys-Dietz syndrome (LDS), but pediatric patients with TAA who do not meet diagnostic criteria for these conditions have not been well characterized. There is, however, mounting evidence to indicate that the aortopathy in these patients also has a genetic and developmental basis. The 2010 guidelines from the American Heart Association provide thresholds for prophylactic replacement of the aorta, based primarily on aortic diameter or rapid aortic enlargement.2 In these guidelines, the threshold for surgery is lower for patients with LDS than MFS and is lower for patients with MFS than “nonsyndromic” TAA, but the stratification of patients who do not have a diagnosis of a specific CTD is limited. In the context of increasing recognition of the genetic basis of TAA, there is a need to characterize this heterogeneous patient population with idiopathic TAA in order to develop risk classification paradigms to guide clinical management. Children with TAA may benefit from medical intervention to prevent disease progression, cardiac events, and need for surgery. We have performed deep phenotyping of cardiovascular and noncardiovascular characteristics in a pediatric cohort of patients with TAA who do not have the diagnosis of a specific CTD such as MFS or LDS. We aimed to generate a phenotype-based framework for risk classification and hypothesized that there are noncardiovascular abnormalities associated with TAA severity.
ARB BAV BB CTD CVM FTAAD LDS MFS MPA STJ TAA
Angiotensin II type 1 receptor blocker Bicuspid aortic valve Beta blocker Connective tissue disorder Cardiovascular malformation Familial thoracic aortic aneurysm or dissection Loeys-Dietz syndrome Marfan syndrome Main pulmonary artery Sinotubular junction Thoracic aortic aneurysm
From the 1Division of Cardiology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH; 2Department of Pediatrics and Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, IN; and 3 Division Human Genetics, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH Supported by the American Heart Association (IRG 18830027 [to R.H.]) and the Cincinnati Children’s Hospital Medical Center’s Research Foundation (to R.H.). The authors declare no conflicts of interest. 0022-3476/$ - see front matter. Copyright ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2015.02.042
131
THE JOURNAL OF PEDIATRICS
www.jpeds.com
We additionally hypothesized that beta blocker (BB) or angiotensin II type 1 receptor blocker (ARB) therapy reduces the rate of aortic dilation. Consideration of the patient’s global phenotype may be useful to stratify risk and guide clinical interventions.
Methods This was a retrospective cohort study consisting of review of existing medical records; the local Institutional Review Board approved this study. Patients with TAA age 21 years or younger were identified by review of cardiovascular genetics clinic rosters from July 1, 2010, to December 31, 2012. TAA was defined as aortic root or ascending aorta z-score greater than +2 by transthoracic echocardiography.3 All patients underwent evaluation by a pediatric cardiologist, medical geneticist, and genetic counselor, which included comprehensive physical examination and detailed pedigree ascertainment. The patients evaluated in this clinic constitute a phenotypically and genetically heterogeneous spectrum including syndromic TAA (eg, MFS, LDS, Turner syndrome), familial TAA or dissection (FTAAD), patients with some concern for a CTD, and patients referred for coordinated management of isolated TAA. Patients with syndromic TAA were excluded. The included patients were further classified by a medical geneticist (S.W.) utilizing the systemic score of the revised Ghent nosology for MFS as having no or low (score 0-1), intermediate (score 2-4), or high (score 5-6) suspicion for MFS.4 In addition to excluding patients with the diagnosis of a specific CTD or syndrome associated with TAA, patients with complex cardiovascular malformations (CVMs) associated with aortic dilation (eg, tetralogy of Fallot) or at least moderate aortic valve stenosis or regurgitation were also excluded. Data collected from the electronic medical record included demographics, anthropomorphic data, vital signs, past medical history, family history, physical examination findings including Beighton score (hypermobility), noncardiovascular features of LDS, and systemic score of the revised Ghent nosology (MFS),4,5 medication use, and genetic testing. At each visit the patient was examined for the development of new findings, which were included in these analyses. Echocardiograms performed from May 1, 2003, to December 31, 2012 were reviewed. Data extracted included aorta and main pulmonary artery (MPA) diameters, aortic and mitral valve morphology and function, and presence of additional CVMs. To improve reliability, aortic dimensions were measured in triplicate by a single reader (B.L.) and 10% were re-measured by a second reader (R.H.). Aortic measurements were obtained with 2-dimensional echocardiography in the parasternal long axis in systole, according to consensus guidelines.2,6 Aortic annulus, aortic root, sinotubular junction (STJ), proximal ascending aorta, and MPA diameters were measured from inner edge to inner edge perpendicular to the long axis of the vessel at maximal expansion during systole. Absolute measurements were indexed to 132
Vol. 167, No. 1 body surface area, using the Haycock formula,7 and normalized to generate z-scores based on reference data from Boston Children’s Hospital for annulus and root8,9 and local data for STJ and ascending aorta. Intervals of therapy and doses were recorded. Atenolol was the prescribed BB in all but 1 patient (metoprolol), and losartan was the only ARB prescribed. Patients were divided into 2 groups: no therapy or therapy (BB or ARB). For exploratory subanalysis, the therapy group was subdivided into BB only or ARB only. Patients with reported noncompliance were considered as not receiving therapy. Statistical Analyses Analyses were performed using JMP statistical software package (SAS Institute, Cary, North Carolina). Continuous data were evaluated for distributional qualities. We examined the variability in the triplicate measures of echocardiographic data using coefficient of variation (cardiovascular). When triplicate values exceeded 5% cardiovascular, the 2 most similar measures were selected. Once consistency was evaluated, the maximum value of the retained replicated measure was used. In addition, 10% of studies were independently remeasured by a second reader (R.H.) with strong correlation (R2 > 0.95) between readers. Baseline characteristics of the cohort were described. For continuous variables distributed normally (eg, age), means and SDs are reported. For continuous variables that were non-normally distributed (eg, echocardiogram intervals, revised Ghent systemic score), medians, and IQRs are used. Categorical data are reported as frequencies. To identify differences between treatment groups, we utilized t tests for continuous normally distributed variables, Wilcoxon rank sum for continuous non-normally distributed variables, and contingency tables and c2 goodness of fit tests for categorical variables. To examine longitudinal changes in aortic dimensions from baseline study, repeated measures analysis (mixed models) was used on z-scores. For individuals in the no therapy group, the baseline study was considered the first echocardiogram. For individuals receiving BB or ARB, the baseline study was considered the last echocardiogram prior to therapy initiation. The primary outcome variables were zscores of the aortic root and ascending aorta. Time was the primary fixed effect with the individual as a random effect. Other covariates included age and presence of noncardiovascular or valvar abnormalities. As we considered 11 distinct noncardiovascular or valvar abnormalities, we employed a Bonferroni correction (a = 0.05/11 = 0.0045) for statistical significance of the rate of aortic root change for noncardiovascular and valvar abnormalities. To test whether there were differences in disease progression between untreated and treated individuals prior to the initiation of treatment, an interaction of treatment status and time was included with the analysis restricted to untreated and pretreatment treated individuals. To test whether initiation of treatment changed progression, we performed an analysis in treated individuals with an interaction term between pre-post Landis et al
ORIGINAL ARTICLES
July 2015 treatment and time. No multiple testing correction was applied for evaluating effect of medical therapy.
Results The medical history of each of the 128 pediatric patients with TAA who were evaluated in cardiovascular genetics clinic over the study period was reviewed. After excluding those diagnosed with a genetic syndrome associated with TAA (n = 48), complex CVM (n = 7), or worse than mild aortic valve stenosis or regurgitation (n = 4), there were 69 (53%) patients who comprised the study population (Figure 1). Referral indications for the study population were evaluation and management of suspected CTD or other genetic disorder in 45 (65%), evaluation and management of TAA without referring caregiver having clear concern for CTD in 18 (26%), and family history of TAA or dissection in 6 (9%) patients. There was a median of 2 (range 1-5) clinic visits and median of 4 (range 1-18) echocardiograms per subject. The study population consisted predominantly of white males in late childhood to adolescence (Table I). The population was slender with mean body mass index of 18 kg/m2 and included 15 (22%) patients exceeding 95th percentile for height. The median systemic score according to the revised Ghent nosology was 2 (IQR 1-4), which is below the diagnostic threshold of 7. Based upon individual systemic scores, there were 30 (43%) patients with low or no suspicion for MFS (score 0 or 1), 31 (45%) with intermediate suspicion for MFS (score 2-4), and 8 (12%) with high suspicion for MFS (score 5 or 6) (Figure 1 and Table II; Table II is available at www.jpeds.com). Although MFS was highly suspected in only a minority of the cohort,
Table I. Patient characteristics, cardiovascular, and noncardiovascular findings according to therapy group Patient characteristics
n (%) No therapy BB or ARB (N = 69) (N = 28) (N = 41)
Demographics Age at initial visit (y) 12.5 (5.3) Sex Male 60 (87) Female 9 (13) Race White 60 (87) Black 6 (9) Asian 1 (1) Other 2 (3) Ethnicity* Non-Hispanic 66 (97) Hispanic 3 (3) Anthropomorphic data, mean (SD) Height percentile 58 (33) Weight percentile 45 (30) BMI absolute 18 (2.6) BMI percentile 35 (26) BSA absolute 1.4 (0.5) Blood pressure, mean (SD) Systolic percentile 56 (27) Diastolic percentile 60 (23) Noncardiovascular abnormalities Skeletal 45 (65) Craniofacial 37 (54) Ocular 22 (32) Neurodevelopmental 17 (25) Cutaneous 13 (19) Gastrointestinal 7 (10) Pulmonary 3 (3) Cardiovascular abnormalities Aortic root z-score +2 to 3 22 (32) z-score +3 to 4 21 (30) z-score > +4 21 (30) Ascending aorta† z-score +2 to 3 18 (29) z-score +3 to 4 7 (11) z-score > +4 12 (19) BAV 17 (25) MVP 11 (16)
P
11.9 (5.9)
12.9 (4.9)
.51
21 (75) 7 (25)
39 (95) 2 (5)
.03
25 (89) 2 (7) 0 1 (4)
35 (85) 4 (10) 1 (2) 1 (2)
.73
27 (100) 0
39 (95) 2 (5)
.51
52 (35) 40 (32) 18 (2.5) 32 (26) 1.3 (0.5)
61 (32) 49 (29) 18 (2.6) 37 (27) 1.4 (0.4)
.27 .23 .74 .48 .48
55 (23) 61 (21)
57 (29) 60 (25)
.7 .9
16 (57) 16 (57) 9 (32) 7 (25) 3 (11) 2 (7) 2 (7)
29 (71) 21 (51) 13 (32) 10 (24) 10 (24) 5 (12) 1 (2)
.25 .63 .97 .95 .21 .69 .56
12 (43) 6 (21) 5 (18)
10 (24) 15 (37) 16 (39)
.18
6 (23) 1 (4) 4 (15) 7 (25) 4 (14)
12 (32) 6 (16) 8 (22) 10 (24) 7 (17)
.71 .95 .76
BMI, body mass index; BSA, body surface area; MVP, mitral valve prolapse. P represents comparison of no therapy vs BB or ARB. *One ethnicity not known. †Unable to measure ascending aorta dimension for 6 patients.
Figure 1. Distribution of patients with TAA in cardiovascular genetics clinic population. The clinic patient population with TAA (N = 128) was comprised of patients with syndromic TAA, CVM associated with TAA, moderate to severe aortic valve dysfunction, and the study cohort (n = 69). The study cohort (white) is subdivided with dashed lines into categories of high (H), intermediate (I), and low/no (L) suspicion for MFS AV, aortic valve.
noncardiovascular abnormalities were frequently observed even among patients who were not referred for a suspected CTD. Altogether, more than 90% of patients were observed to have at least 1 noncardiovascular abnormality, including 45 (65%) patients with skeletal abnormalities, 37 (54%) with craniofacial abnormalities, and 22 (32%) with ocular abnormalities (Tables I and III; Table III is available at www.jpeds.com). Joint hypermobility, defined as Beighton score of 5 or greater, was infrequently observed (n = 7, 10%); none met clinical criteria for hypermobility EhlersDanlos syndrome. Among the study cohort, FBN1, TGFBR1, and TGFBR2 sequencing was negative for pathogenic variants in 42 (61%) patients, and additional deletion/duplication testing was negative in 3 patients (Table II). The majority (77%)
Clinical Stratification of Pediatric Patients with Idiopathic Thoracic Aortic Aneurysm
133
THE JOURNAL OF PEDIATRICS
www.jpeds.com
of patients in the study cohort who were determined to have intermediate or high suspicion for MFS had negative testing for FBN1, TGFBR1, and TGFBR2 mutations. At the time of last evaluation, 8 (12%) patients had a diagnosis of FTAAD based on confirmed family history of autosomal dominant TAA. Sequencing of ACTA2 and MYH11 was normal in 15 (22%) of the entire cohort. None of those with high suspicion for MFS or LDS was tested for ACTA2 and MYH11 mutations compared with 25% in the intermediate or low/no suspicion groups. These patterns indicate that over the study period, genetic testing was performed in a tiered manner based on suspicion for MFS or other CTD. Overall, the high frequency of noncardiovascular phenotype abnormalities in this collectively heterogeneous study cohort indicates that many pediatric patients with TAA have some degree of nonspecific systemic connective tissue involvement. The aortic root was dilated in 64 (93%) patients, but ascending aorta was dilated in 37 (59%) and annulus in only 3 patients (Table I). The mean z-score at baseline measurement was 3.1 1.1 (2.9 0.7 cm) for the aortic root and 2.0 1.7 (2.5 0.6 cm) for the ascending aorta (Figure 2). There were no cases of aortic dissection or surgical aortic replacement over the study period. Bicuspid aortic valve (BAV) was present in 17 (25%) patients, among whom 76% had fusion of the right and left coronary cusps, which is consistent with prior observations.10-12 Notably, 15 (88%) of the patients with BAV had aortic root dilation. In 11 of these cases of BAV, there was associated dilation of the ascending aorta; 2 patients had dilation restricted to the ascending aorta. Noncardiovascular abnormalities were also frequent in the BAV subset and similar to the overall cohort: 41% with skeletal, 53% with craniofacial, and 29% with ocular
Vol. 167, No. 1 abnormalities. The MPA was dilated in 19 (28%) patients, although image quality precluded measurement in a significant proportion of studies (43%), suggesting this may be an underestimate. All patients had normal left ventricular systolic function. Using longitudinal modeling (repeated measures), we observed significant associations between the rate of aortic enlargement and noncardiovascular abnormalities (Table IV). For example, ocular (P = .002) or cutaneous (P = .003) abnormalities were associated with an increased rate of aortic root dilation. Meanwhile, craniofacial (P < .001) abnormalities were associated with an increased rate of ascending aorta dilation. The severity of craniofacial dysmorphism, considered as the sum of craniofacial abnormalities in an individual, was also associated with rate of ascending aorta dilation (P < .001). There was no significant association between systemic score and rate of aortic enlargement or between BAV or mitral valve prolapse and the rate of aortic enlargement. When the cohort subset who had normal sequencing of FBN1, TGFBR1, and TGFBR2 were independently analyzed, the associations for cutaneous and craniofacial abnormalities with TAA progression remained significant, but ocular findings were no longer significant (Table V; available at www.jpeds.com). These observations provide a basis for stratifying the risk of TAA progression according to the pattern of a patient’s noncardiovascular abnormalities. BB or ARB therapy was initiated in 41 (59%) patients, including 24 (35%) on BB and 17 (25%) on ARB therapy. The mean age at therapy initiation was 12.9 4.9 years. The mean daily dose at the last follow up was 0.7 0.4 mg/kg for atenolol and 0.7 0.3 mg/kg for losartan (Table VI; available at www.jpeds.com). Although most patients were normotensive (Table I), 3 had systemic
Figure 2. A, Two-dimensional echocardiographic image of the proximal aorta. The aortic segments of interest are indicated by the double arrowhead lines and include: (1) aortic annulus; (2) aortic root; (3) STJ; and (4) ascending aorta. B, Bar plot demonstrating that the aortic root is more severely dilated at baseline compared with other aortic segments, and at baseline echocardiogram the patients who receive therapy have more severe aortic dilation than patients not receiving therapy. Bars represent mean z-score and error plots represent SDs from the mean. LA, left atrium; LV, left ventricle. 134
Landis et al
ORIGINAL ARTICLES
July 2015
Table IV. Noncardiovascular and cardiovascular factors impacting rate of aortic dilation
Affected systems Skeletal Craniofacial Craniofacial (sum) Ocular Neurodevelopmental Cutaneous Gastrointestinal Pulmonary Height >95th percentile Systemic score Cardiovascular BAV MVP
Relative change in rate of aortic root dilation 4 4 4 [[ 4 [[ 4 4 4 4 4 4
P
Relative change in rate of ascending aorta dilation
P
.4 .15 .54 .002 .88 .003 .10 .53 .37 .09
[ [[ [[ 4 4 Y 4 4 [ 4
.04 95th percentile Arm span/height >1.05 Low upper segment: lower segment (per Ghent criteria) Joint pain Pectus carinatum Decreased elbow extension Syndactyly Contractures Craniofacial Dolichocephaly Enophthalmos Downslanting palpebral fissures Malar hypoplasia Retrognathia Hypertelorism Palate abnormality Orthodontia Uvula abnormality Ear malformation Ocular Myopia Pupil/extraocular muscle abnormality Cutaneous Striae Easy bruising Abnormal scarring Gastrointestinal Inguinal hernia Eosinophilic esophagitis Pulmonary (pneumothorax) Cardiovascular abnormalities Aortic root z-score +2 to 3 z-score +3 to 4 z-score > +4
n (% of total)
No therapy (N = 28)
BB or ARB (N = 41)
BB (N = 24)
ARB (N = 17)
Pz
Px
12.5 (5.3)
11.9 (5.9)
12.9 (4.9)
13.1 (4.9)
12.5 (5.1)
.51 .026
.71 .17
60 (87) 9 (13)
21 (75) 7 (25)
39 (95) 2 (5)
24 (100) 0
15 (88) 2 (12) .73
.27
60 (87) 6 (9) 1 (1) 2 (3)
25 (89) 2 (7) 0 1 (4)
35 (85) 4 (10) 1 (2) 1 (2)
22 (92) 2 (8) 0 0
13 (76) 2 (12) 1 (6) 1 (6)
66 (97) 3 (3)
27 (100) 0
39 (95) 2 (5)
23 (96) 1 (4)
16 (94) 1 (6)
58 (33) 45 (30) 18 (2.6) 35 (26) 1.4 (0.5)
52 (35) 40 (32) 18 (2.5) 32 (26) 1.3 (0.5)
61 (32) 49 (29) 18 (2.6) 37 (27) 1.4 (0.4)
63 (32) 47 (30) 18 (2.2) 33 (27) 1.4 (0.4)
59 (33) 50 (28) 18 (3.1) 42 (27) 1.4 (0.5)
.27 .23 .74 .48 .48
.69 .76 .62 .27 .75
56 (27) 60 (23)
55 (23) 61 (21)
57 (29) 60 (25)
61 (33) 65 (23)
54 (25) 54 (27)
.7 .9
.49 .24
45 (65) 20 (29) 14 (20) 13 (19) 12 (17) 13 (19) 5 (7) 8 (12) 15 (22) 5 (7) 3 (4)
16 (57) 6 (21) 6 (21) 4 (14) 4 (14) 4 (14) 3 (11) 3 (11) 5 (18) 3 (11) 1 (4)
29 (71) 14 (34) 8 (20) 9 (22) 8 (20) 9 (22) 2 (5) 5 (12) 10 (24) 2 (5) 2 (5)
18 (75) 7 (29) 4 (17) 5 (21) 4 (17) 6 (25) 2 (8) 4 (17) 7 (29) 0 1 (4)
11 (65) 7 (41) 4 (24) 4 (24) 4 (24) 3 (17) 0 1 (6) 3 (17) 2 (12) 1 (6)
.25
.51
5 (7) 3 (4) 4 (6) 4 (6) 2 (3) 37 (54) 7 (10) 1 (1) 9 (13) 23 (33) 8 (12) 7 (10) 9 (13) 7 (10) 2 (3) 8 (12) 22 (32) 17 (25) 5 (7) 13 (19) 13 (19) 0 0 7 (10) 4 (6) 4 (6) 3 (4)
1 (4) 0 1 (4) 2 (7) 0 16 (57) 4 (14) 0 4 (14) 12 (43) 3 (11) 3 (11) 4 (14) 3 (11) 2 (7) 4 (14) 9 (32) 8 (29) 1 (4) 3 (11) 3 (11) 0 0 2 (7) 1 (4) 2 (7) 2 (7)
4 (10) 3 (7) 3 (7) 2 (5) 2 (5) 21 (51) 3 (7) 1 (2) 5 (12) 11 (27) 5 (12) 4 (10) 5 (12) 4 (10) 0 4 (10) 13 (32) 9 (22) 4 (10) 10 (24) 10 (24) 0 0 5 (12) 3 (7) 2 (5) 1 (2)
3 (13) 1 (4) 2 (8) 2 (8) 2 (8) 13 (54) 2 (8) 1 (4) 3 (13) 7 (29) 3 (13) 3 (13) 3 (13) 1 (4) 0 2 (8) 7 (29) 5 (21) 2 (8) 7 (29) 7 (29) 0 0 3 (12) 2 (8) 1 (4) 1 (4)
1 (6) 2 (12) 1 (6) 0 0 8 (47) 1 (6) 0 2 (12) 4 (24) 2 (12) 1 (6) 2 (12) 3 (18) 0 2 (12) 6 (35) 4 (24) 2 (12) 3 (18) 3 (18) 0 0 2 (12) 1 (6) 1 (6) 0
.63
.75
.97
.74
.21
.48
22 (32) 21 (30) 21 (30)
12 (43) 6 (21) 5 (18)
10 (24) 15 (37) 16 (39)
4 (17) 8 (33) 12 (50)
6 (35) 7 (41) 4 (24)
.51
Clinical Stratification of Pediatric Patients with Idiopathic Thoracic Aortic Aneurysm
1.0
.69
1.0
.56
1.0
.18 .11 (continued )
137.e4
THE JOURNAL OF PEDIATRICS
www.jpeds.com
Vol. 167, No. 1
Table III. Continued Patient characteristics
n (% of total)
No therapy (N = 28)
BB or ARB (N = 41)
BB (N = 24)
ARB (N = 17)
Pz
Px
.71 .95 .76
1.0 .79 .61
†
Ascending aorta z-score +2 to 3 z-score +3 to 4 z-score > +4 BAV MVP
18 (29) 7 (11) 12 (19) 17 (25) 11 (16)
6 (23) 1 (4) 4 (15) 7 (25) 4 (14)
12 (32) 6 (16) 8 (22) 10 (24) 7 (17)
7 (33) 4 (19) 5 (23) 5 (21) 3 (13)
5 (31) 2 (13) 3 (19) 5 (29) 4 (24)
BMI, body mass index; BSA, body surface area. *One ethnicity not known. †Unable to measure ascending aorta dimension for 6 patients. zNo therapy vs BB or ARB. xBB vs ARB.
Table VI. Therapy dosage of individual patients BB Age therapy started (y)
Table V. Non-cardiovascular and cardiovascular factors impacting rate of aortic dilation only among patients with negative testing for FBN1, TGFBR1, and TGFBR2
Affected systems Skeletal Craniofacial Craniofacial (sum) Ocular Neurodevelopmental Cutaneous Gastrointestinal Pulmonary Height >95th percentile Systemic score Cardiovascular BAV MVP
P
Relative change in rate of ascending aorta dilation
P
4 4 4 [ 4 [[ Y 4 4 4
.54 .96 .41 .03 .12
