Young patients with congenital aortic valve disease are at risk of left ventricular (LV) diastolic dysfunction (DD). We evaluated LV remodeling and the prevalence of, and risk factors for, DD in patients with aortic stenosis (AS), pure aortic regurgitation (AR), and AS+AR. Patients aged 8 to 39 years with congenital AS (n = 103), AR (n = 36), or AS+AR (n = 107) were identified. Cross-sectional assessment of the LV remodeling pattern and diastolic function was performed. A diastolic function score (DFS; range 0 to 4) was assigned to each patient, with 1 point for an abnormal value in each of 4 categories: mitral inflow (E/A and E-wave deceleration time), tissue Doppler E′, E/E′, and left atrial volume. Patients with a DFS of ≥2 were compared to those with a DFS <2. Concentric hypertrophy was the most common remodeling pattern in those with AS (51%), mixed/physiologic hypertrophy in those with AS+AR (48%) and eccentric hypertrophy in those with AR (49%) predominated. In the entire cohort, 91 patients (37%) had a DFS of ≥2. Patients with AS or AS+AR had greater DFS than those with pure AR (p <0.001). On multivariate analysis, a greater LV mass z-score and previous aortic valve balloon dilation were associated with a DFS of ≥2. In patients with catheterization data (n = 65), E/E′ correlated with LV end-diastolic pressure. Those with a DFS of ≥2 had a greater LV end-diastolic pressure and mean pulmonary artery pressure than those with a DFS <2. In conclusion, DD is common in young patients with AS and AS+AR but not in those with pure AR. A greater LV mass and previous aortic valve dilation were associated with DD.
The effect of chronic pressure and volume loading due to aortic valve disease on left ventricular (LV) remodeling and compliance has been well described in adults. Chronic pressure loading leads to LV remodeling with the development of concentric hypertrophy. Early in the disease course, concentric hypertrophy allows the wall stress to remain normal and allows preservation of systolic function. Later, the deleterious effects of concentric hypertrophy and associated myocardial fibrosis become apparent, with the development of systolic and diastolic dysfunction (DD). The myocardial response to chronic pressure load due to congenital aortic valve disease in children is also characterized by concentric hypertrophy, myocardial fibrosis, and impaired diastolic function. The time course and risk factors for progression of DD might be different in younger patients with congenital aortic stenosis (AS). The effect of LV volume load due to aortic regurgitation (AR) on diastolic function is less clear, with most adult data showing a high incidence of DD, but experimental and pediatric data showing less of an effect of volume loading on diastolic function. The effect of chronic combined pressure and volume load due to AS+AR on diastolic function in younger patients has not been described. In the present cross-sectional study of children and young adults with congenital aortic valve disease, we describe the LV remodeling pattern and prevalence of, and risk factors for, DD in patients with AS, pure AR, and AS+AR.
Methods
The records of all patients age 8 to 39 years evaluated at our institution from January 2005 to May 2011 with moderate or greater congenital AS and/or more than mild AR were retrospectively reviewed. The exclusion criteria included congenital heart disease (with the exception of bicommissural aortic valve and aortic coarctation), previous cardiac surgery with cardiopulmonary bypass, residual aortic arch obstruction (gradient >20 mm Hg), systemic hypertension, chronic renal disease, acquired valve disease, orthotopic heart transplantation, a history of diseases or therapies known to affect diastolic function (coronary artery disease, chemotherapy, Kawasaki disease). Baseline demographics, clinical characteristics, and clinical course, including cardiac interventions, were collected.
The patients were classified into 1 of 3 groups according to the predominant aortic valve disease: AS, pure AR, or AR+AR. The AS group included patients with both moderate or greater AS and mild or less AR. The AS+AR cohort consisted of patients with both greater than mild AR and moderate or greater AS. The pure AR cohort included patients with more than mild AR and no AS (AS gradient <15 mm Hg and no history of balloon aortic valvuloplasty for congenital AS). We defined moderate or greater AS by a Doppler gradient of ≥36 mm Hg using the greater value of the maximum instantaneous gradient from the apical imaging window or mean gradient from the suprasternal notch window and/or a history of balloon aortic valvuloplasty. Qualitatively grading of AR in our echocardiographic laboratory was performed using a 4-point ordinal scale (0, none; 1, trivial; 2, mild; 3, moderate; and 4, severe) by ½-unit increments and determined by a combination of previously published criteria. AR was considered more than mild if ≥1 of the following criteria were met: pandiastolic flow reversal in the descending abdominal aorta, vena contracta width/body surface area >3.1 mm/m 2 , LV end-diastolic volume z-score >2. The Committee for Clinical Investigation at Children’s Hospital Boston approved the use of the patient medical records for this review.
The most recent complete echocardiogram that included a full interrogation of diastolic function was reviewed. The AS gradient and AR grade were collected from reports produced at the time of the study. The following LV parameters were recorded: end-diastolic volume, mass, mass:volume, and ejection fraction and the z-scores for these variables. The LV end-diastolic volume was calculated using the 5/6 area-length formula and LV mass using volumetric 2-dimensional measurements. The pattern of LV remodeling was classified according to previously established criteria : normal ventricle (normal mass, volume, and mass:volume), concentric remodeling/hypertrophy (normal LV volume, high LV mass, and/or mass:volume), eccentric remodeling/hypertrophy (high volume, normal mass, and low or normal mass:volume), or mixed/physiologic hypertrophy (high mass, high volume with normal or high mass:volume).
All measurements of diastolic variables were retrospectively remeasured by a single echocardiographer (K.F.) from images obtained at the echocardiogram. Standard mitral valve inflow pulsed-Doppler indexes of diastolic function, including peak early (E) and late (A) diastolic transmitral velocities, E/A, and E-wave deceleration time, were measured. Pulsed wave tissue Doppler imaging velocities were obtained from the lateral mitral annulus and the interventricular septum from the apical 4-chamber view. Only tracings that demonstrated a clear E′ were used. Each tissue Doppler imaging velocity was measured on 3 consecutive cardiac cycles, and the average value was used. The peak early mitral inflow velocity/early mitral tissue Doppler imaging velocity (E/E′) was calculated. The left atrial volumes were calculated using the prolate-ellipse formula. A left atrial volume ≥32 ml/m 2 was considered abnormal. For all other diastolic function variables, the z-scores derived from normative data at our institution using a previously described technique were used, and a z-score of >2 or <−2 was considered abnormal. Examinations were performed using commercially available ultrasound equipment (Philips iE33, Koninklijke Philips Electronics, NV, Amsterdam, The Netherlands).
The diastolic parameters were grouped into 1 of 4 categories for analysis: (1) pulsed-wave Doppler mitral inflow (E/A, E-wave deceleration time), (2) tissue Doppler imaging velocities (mitral annular and septal E′), (3) E/E′, and (4) left atrial volume. The patients were assigned a diastolic function score (DFS) of 0 to 4, with 1 point for an abnormal value in each category.
For patients who underwent catheterization within 3 months of the echocardiogram, the hemodynamic data were collected from reports produced at the catheterization (n = 65). For cases in which interventions were performed (e.g., balloon aortic valvuloplasty), the preintervention hemodynamic data were included in the analysis.
The demographic and clinical and testing data are reported as counts for categorical variables and as the median and interquartile range for continuous variables. The comparisons of the demographic, clinical, and echocardiographic data among the patients with AS, AS+AR, and AR were made using Fisher’s exact test for categorical variables and the Kruskal-Wallis test for continuous variables. To evaluate the risk factors for DD, patients with a DFS <2 were compared to those with a DFS ≥2. The associations between the demographic, clinical, and echocardiographic risk factors and a DFS ≥2 were assessed. Multivariate analysis with stepwise logistic regression was used to assess for factors associated with DFS of ≥2.
For the subset of patients with catheterization data, the associations between echocardiographic markers of left heart filling pressures (E/E′), and invasively measured hemodynamic data were evaluated using Pearson’s correlation coefficients. Receiver operating characteristic curves were constructed to assess the ability of E/E′ to predict elevated LV end-diastolic pressure. All statistical analyses were 2-sided, and a type I error was controlled at a level of 0.05. Analyses were performed with SPSS, version 16.0 (SPSS, Chicago, Illinois).
Results
The cohort consisted of 246 patients: 103 with AS, 107 with AS+AR, and 36 with pure AR. Patients with AS and AS+AR were older (p = 0.003) and were more likely to have undergone balloon aortic valvuloplasty than those with pure AR (p <0.001; Table 1 ).
| All Patients (n = 246) | AS (n = 103) | AS+AR (n = 107) | AR (n = 36) | p Value | |
|---|---|---|---|---|---|
| Age (yrs) | 17 (14–21) | 18 (14–23) | 16.4 (14–21) | 14.7 (10–18) | 0.003 † |
| Age >20 yrs | 171 (70%) | 68 (66%) | 74 (69%) | 7 (19%) | 0.025 † |
| Weight (kg) | 61 (45–76) | 61 (48–73) | 65 (46–81) | 51 (36–65) | 0.022 † |
| Body surface area (m 2 ) | 1.7 (1.4–1.9) | 1.7 (1.5–1.9) | 1.8 (1.4–2.0) | 1.5 (1.2–1.8) | 0.93 |
| Male gender | 191 (78%) | 73 (71%) | 89 (83%) | 29 (81%) | 0.068 ∗ |
| Previous balloon aortic valvuloplasty | 111 (45%) | 49 (48%) | 62 (58%) | 0 (0%) | <0.001 † |
| Intervention within first year of life | 44 (18%) | 21 (20%) | 23 (21%) | 0 (0%) | 0.011 † |
| Repeat aortic valvuloplasty | 27 (11%) | 9 (9%) | 18 (17%) | 0 (0%) | 0.096 † |
| Unicommissural aortic valve | 71 (29%) | 34 (33%) | 31 (32%) | 6 (17%) | 0.10 |
| Systolic blood pressure | 111 (101–121) | 108 (99–118) | 114 (105–122) | 107 (102–122) | 0.14 |
| Diastolic blood pressure | 64 (56–73) | 67 (59–77) | 63 (53–69) | 58 (54-63) | <0.001 ∗† |
∗ Significant difference between AS and AS+AR groups (p <0.05).
† Significant difference between AS+AR and AR groups (p <0.05).
Most patients with AS had concentric hypertrophy (51%) or normal ventricle (39%; Figure 1 ). Patients with AS+AR disease most commonly had mixed/physiologic hypertrophy (48%) or concentric hypertrophy (25%), while in those with pure AR, eccentric hypertrophy (49%) and a normal ventricle (32%) predominated.
The LV end-diastolic volume was greater in those with AS+AR and AR than in those with AS (p <0.001; Table 2 ). The LV mass z-score was greatest in the patients with AS+AR followed by those with AR and then patients with AS (p <0.001). However, the LV mass:volume was greatest in patients with AS, intermediate in those with AS+AR, and lowest in those with AR (p <0.001).
| All Patients (n = 246) | AS (n = 103) | AS+AR (n = 107) | AR (n = 36) | p Value | |
|---|---|---|---|---|---|
| Left ventricular end-diastolic volume z-score | 1.2 (−0.3–3.2) | −0.3 (−1.1–0.9) | 2.4 (0.6–4.1) | 3.2 (1.5–7.1) | <0.001 ∗† |
| Left ventricular mass z-score | 1.9 (0.8–3.4) | 1.1 (0.1–2.15) | 2.9 (1.4–4.2) | 1.8 (0.9–3.9) | <0.001 ∗ |
| Left ventricular mass:volume | 1.0 (0.8–1.2) | 1.1 (0.9–1.3) | 0.9 (0.8–1.0) | 0.8 (0.7–0.9) | <0.001 ∗† |
| Left ventricular mass:volume z-score | 1.0 (−0.5–2.2) | 1.7 (0.3–3.5) | 0.7 (−0.7–1.5) | −0.82 (−1.8–0.1) | <0.001 ∗† |
| Aortic regurgitation grade | 2.5 (2–3) | 2 (0–2) | 3 (2.5–3.0) | 3 (2.5–4) | <0.001 ∗ |
| Aortic stenosis gradient (mm Hg) | 36 (25–52) | 45 (30–57) | 40 (30–50) | 12 (0–17) | <0.001 † |
| Left ventricular ejection fraction (%) | 67 (62–71) | 66 (62–71) | 67 (61–72) | 66 (61–73) | 0.06 |
∗ Significant difference between AS and AS+AR groups (p <0.01).
† Significant difference between AS+AR and AR groups (p <0.01).
In the entire cohort, 186 patients (73%) had a DFS of ≥1 and 91 (37%) had a DFS ≥2 ( Figure 2 ). The percentage of patients with abnormal diastolic indexes was similar between the AS+AR and AS and both groups had a greater DFS than the AR group (p <0.001 for a DFS of ≥1 and ≥2).
The left atrial volume and pulsed Doppler mitral inflow parameters did not vary among the groups, although significant differences in the tissue Doppler imaging value and E/E′ were present ( Table 3 ). The mitral annular and septal E′ were similar between patients with AS and AS+AR and were lower than those with AR (p <0.001). The E/E′ values and the percentage of patients with an E/E′ z-score of ≥2 were similar between those with AS and AS+AR and greater than in those with AR (p <0.001).
| All Patients (n = 246) | AS (n = 103) | AS+AR (n = 107) | AR (n = 36) | p Value | |
|---|---|---|---|---|---|
| Mitral inflow E/A | 2.0 (1.6–2.5) | 1.9 (1.5–2.4) | 2.1 (1.6–2.6) | 2.1 (1.7–2.6) | 0.29 |
| Mitral inflow E/A z-score | −0.4 (−1.2–0.3) | −0.6 (−1.4–0.1) | −0.5 (−1.2–0.4) | −0.1 (−0.9–0.5) | 0.12 |
| Mitral inflow deceleration time (ms) | 153 (127–176) | 154 (128–181) | 148 (120–174) | 153 (131–186) | 0.34 |
| Mitral inflow deceleration time z-score | −0.5 (−1.1–0.3) | −0.5 (1.0–0.1) | −0.5 (−1.3–0.6) | −0.1 (−0.7–0.5) | 0.12 |
| Mitral annular E′ (cm/s) | 13.5 (11.1–16.0) | 13.0 (10.5–15.5) | 13.1 (10.7–15.2) | 16.6 (13.9–17.9) | <0.001 ∗ |
| Mitral annular E′ z-score | −1.8 (−2.6 to −1.0) | −2.0 (−4.9–1.3) | −2.1 (−2.8 to −1.2) | −0.6 (−1.4 to −0.03) | <0.001 ∗ |
| Mitral annular E′ z-score ≤−2 | 107 (44%) | 48 (46%) | 58 (54%) | 1 (3%) | <0.001 ∗ |
| Septal E′ (cm/s) | 10.0 (8.7–11.7) | 10.2 (9.0–11.7) | 9.7 (8.2–11.0) | 12.2 (10.8–13.3) | <0.001 ∗ |
| Septal E′ z-score | −1.7 (−2.3 to −1.0) | −1.7 (−2.2 to −1.0) | −2.0 (−2.5 to −1.3) | −0.6 (−1.3 to −0.1) | <0.001 ∗ |
| Septal E′ z-score ≤−2 | 92 (37%) | 37 (36%) | 52 (49%) | 3 (8%) | <0.001 ∗ |
| E/E′ | 8.2 (6.7–10.0) | 8.6 (7.2–10.0) | 8.7 (7.0–10.5) | 6.3 (5.5–6.9) | <0.001 ∗ |
| E/E′ z-score ≥2 | 69 (28%) | 29 (28%) | 38 (36%) | 2 (6%) | 0.003 ∗ |
| Left atrial volume (ml/m 2 ) | 20 (16–24) | 21 (15–25) | 20 (16–24) | 19 (16–23) | 0.62 |
| Left atrial volume >32 ml/m 2 | 54 (22%) | 19 (18%) | 28 (26%) | 7 (19%) | 0.75 |
| 4-Point diastolic score ≥1 | 176 (72%) | 73 (71%) | 85 (80%) | 18 (50%) | 0.006 ∗ |
| 4-Point diastolic score ≥2 | 91 (37%) | 38 (37%) | 50 (47%) | 3 (8%) | <0.001 ∗ |
∗ Significant difference between AS+AR and AR groups (p <0.05).
In the AS group, concentric hypertrophy/remodeling was associated with a DFS ≥2 compared to mixed/physiologic hypertrophy or a normal ventricle (50%, 30%, and 15%, respectively, p <0.001). In the AS+AR and AR remodeling pattern was not associated with the DFS. Subgroup analysis of patients with AS comparing patients with mild or less residual AS (n = 37) and those with more than residual mild AS (n = 66) showed a lower LV mass z-score (median 0.84, interquartile range −1.8 to 3.8 vs median 1.52, interquartile range −0.8 to 10.2, p = 0.026) and lower LV mass:volume (median 0.9, interquartile range 0.7 to 1.6, vs median 1.2, interquartile range 0.8 to 1.8, p = 0.001) in patients with lower gradients. However, no difference was found in the demographics, cardiac interventions, or diastolic function parameters.
The results of univariate analysis of factors associated with a DFS of ≥2 is listed in Table 4 . On multivariate analysis, only a greater LV mass z-score and previous balloon aortic valvuloplasty were associated with a DFS of ≥2.
