In adults with idiopathic dilated cardiomyopathy (IDC), mitral regurgitation (MR) is associated with adverse prognosis and is often addressed by surgery or intervention. MR is commonly found in children with IDC, but its prognostic relevance has not been defined, and interventions to reduce MR are not routinely performed in this population. In this study, it was hypothesized that MR is an independent risk factor for death or transplantation. This was a single-center, retrospective study of sequential patients with IDC or familial IDC (left ventricular end-diastolic dimension z score >2 and ejection fraction <50%). Patients with acute myocarditis or previous mitral surgery were excluded. MR severity was graded according to American Society of Echocardiography guidelines as mild, moderate, or severe on the basis of MR jet vena contracta width. Left ventricular end-diastolic volume, end-systolic volume, and ejection fraction were measured by biplane Simpson’s method. Forty-two children with IDC were studied. The mean follow-up period was 25 months. At initial assessment, 34 children (82%) were taking angiotensin-converting enzyme inhibitors, 25 (60%) furosemide, 27 (65%) β blockers, and 7 (17%) intravenous inotropes. The mean indexed end-systolic volume was 91 ± 51 ml/m 2 . The mean ejection fraction was 27 ± 16%. MR was mild in 42%, moderate in 19%, severe in 2%, and absent in 35% of patients. MR severity progressed from initial to last evaluation. MR severity was an independent risk factor for lower freedom from death or transplantation. Progression in MR severity increased the annual hazard of death or transplantation by a factor of 2.4 (p = 0.003). In conclusion, MR severity is independently associated with worse clinical status and decreased freedom from death or transplantation in children with IDC.
In children with idiopathic dilated cardiomyopathy (IDC), functional mitral regurgitation (MR) is commonly detected by echocardiography. However, functional MR has not been reported as a risk factor for adverse outcomes in children with IDC, and its significance is poorly defined. Likewise, interventions to reduce functional MR in children with IDC are performed less than in adults. We hypothesized that the presence of functional MR and its severity are associated with worse functional status and decreased freedom from death or listing for heart transplantation in children with IDC. Accordingly, the aim of this study was to investigate the relation of functional MR to functional status and death or heart transplantation in children with IDC.
Methods
We retrospectively studied all children with IDC (including syndrome-associated or familial) listed in the heart failure database at our institution from 2004 and 2009. Follow-up data were through February 2010. We chose this time frame because a dedicated and specialized heart failure service with a systematic cardiomyopathy database and standardized approach was established during this period, allowing reliable documentation of patient evaluation and outcomes using a standardized clinical and echocardiographic approach. IDC was diagnosed when clinical symptoms of heart failure were present with echocardiographic findings of a left ventricular (LV) end-diastolic dimension z score >2 (on the basis of institutional z scores) and an ejection fraction (EF) <50%. To exclude patients with acute dysfunction and to study a relatively homogenous population, patients were excluded if they had received anthracyclines, had congenital structural heart disease, had undergone mitral surgery, or were diagnosed with acute myocarditis, LV noncompaction, hypertrophic cardiomyopathy, or LV dysfunction after cardiac arrest.
All patients underwent comprehensive echocardiography following a standardized protocol. To assess the prognostic value of MR at presentation and then later during the clinical course, 2 echocardiograms were analyzed for each patient: the first obtained at our institution, typically at the time of diagnosis, and the second obtained before the primary clinical outcome of death or listing for cardiac transplantation or last follow-up if the patient had not reached either of these end points. There were too few intermediate echocardiographic studies with a long enough time interval between them (6 to 12 months) for further serial analysis. Echocardiographic images were acquired from standard apical and parasternal windows, with patients in the left lateral decubitus position. Because breath holding is impractical in many children, images were acquired during quiet respiration. Our laboratory uses standardized protocols, and studies are performed by trained pediatric echocardiographic sonographers well versed in the protocols. Transducers appropriate for patient size were used. Ultrasound machine settings were standard across the laboratory, with adjustment by the sonographer to optimize the image during the examination. Color gain was set to a level below which random color pixels appear in the tissue. The color scale was generally set at the highest limit obtainable by the transducer frequency and imaging depth. To increase frame rates, the color box was adjusted to cover only the area of interest. Quantitative measurements were made offline using commercially available software (Syngo Dynamics; Siemens Medical Systems, Malvern, Pennsylvania).
All echocardiographic indexes were remeasured for the purposes of the study by a single investigator (F.P.F.) from the original digitally stored images (SyngoDx; Siemens Medical Solutions USA, Inc., Mountain View, California). LV end-diastolic and end-systolic volumes and EF were measured using the modified biplane Simpson method. Semiquantitative grading of MR severity was based on the guidelines of the American Society of Echocardiography. The vena contracta width of the mitral regurgitant jet was measured in the parasternal long-axis and 4-chamber views, and the results were averaged. A vena contracta width <0.3 cm was considered indicative of mild MR, between 0.3 and 0.69 cm of moderate MR, and >0.7 cm of severe MR. Other parameters used to grade MR degree were the jet area, continuous-wave Doppler jet profile, pulsed-wave Doppler transmitral flow, and pulmonary venous flow. Pulsed tissue Doppler velocities were acquired at the lateral and septal mitral annulus, and the peak early diastolic velocity (E′) was measured. The average of the peak E′ velocity at these 2 locations was used to calculate the E/E′ ratio.
Age, gender, age at diagnosis, and medications were recorded from the medical record. All echocardiographic studies were performed on the same day as the clinical examinations. Functional status was graded using the Ross score, which was retrospectively determined on the basis of the respiratory pattern at rest and during feeding or exercise as recorded in the medical record. Because the exact onset of disease cannot be determined, we used the date of the initial echocardiographic exam at the time of IDC diagnosis as the date of diagnosis. Follow-up duration was calculated from date of diagnosis to date of transplantation, death, or last echocardiographic exam at the time of study end.
Demographic, functional, and echocardiographic characteristics and clinical outcomes were compared between patients with and without MR using the unpaired Student’s t test. The primary study outcome was freedom from cardiac transplantation or death. Using Kaplan-Meier nonparametric regression analysis, the presence and degree of MR were tested as independent predictors of the primary outcome. Previously identified risk factors for adverse outcomes in adults and children with IDC, including EF, end-systolic volume, E/E′ ratio, and functional class, were tested as potential confounding factors for the primary outcome. Statistical analysis was performed using SAS version 9.2 (SAS Institute Inc., Cary, North Carolina). The study was approved by the institutional research ethics board.
Results
Forty-two consecutive patients with IDC evaluated at our institution from 1993 to 2010 were enrolled. Table 1 lists clinical characteristics and medical therapy of patients with IDC with and without MR. There were no significant differences between the groups with regard to age, gender, body surface area, or heart failure medications used.
| Variable | MR | |
|---|---|---|
| Yes (n = 26) | No (n = 15) | |
| Age (years) | 7.4 ± 6.9 | 6.6 ± 6.5 |
| Male | 65% | 53% |
| Body surface area (m 2 ) | 0.84 ± 0.6 | 0.66 ± 0.5 |
| Ross score | 2.9 ± 1.1 | 2.3 ± 1.2 |
| Diuretic use | 50% | 36% |
| Angiotensin-converting enzyme inhibitor use | 92% | 96% |
| β blocker use | 60% | 65% |
Echocardiographic characteristics of patients with and without MR are listed in Table 2 . At the initial echocardiographic exam, MR was present in 27 of 42 patients (64%). It was graded as mild in 43%, moderate in 19%, and severe 2%. Overall, the degree of MR progressed from the initial study to last follow-up, when 31% had moderate MR and 8% had severe MR. Patients with MR had higher ventricular volumes and lower EFs than those without MR ( Table 2 ).
| Variable | MR | ||
|---|---|---|---|
| Yes | No | p Value | |
| LV end-diastolic volume index | 139 ± 52 | 77 ± 35 | <0.001 |
| LV end-systolic volume index | 111 ± 48 | 50 ± 32 | <0.001 |
| EF (Simpson) | 19 ± 9 | 39 ± 16 | <0.001 |
| Mitral valve E/E′ ratio | 10.5 ± 6 | 14.2 ± 6 | <0.001 |
The median follow-up time from the initial echocardiographic exam to the last follow-up exam was 9 months, with a large range (1 to 202 months). In the group of patients without MR (n = 15), there were 2 deaths, neither attributable to heart failure. One death occurred in a child with an inborn error of fatty acid oxidation after 1 month of follow-up. The second death occurred in a patient with profound neurologic impairment at 9 months of follow-up. In addition, 3 of 15 patients (13%) were transplanted at 1, 6, and 26 months of follow-up. In the group with MR (n = 26), there was 1 death 34 days after diagnosis. Twelve patients (46%) were transplanted: 4 within 1 month of follow-up, 5 from 6 to 12 months of follow-up, and 2 after 12 months of follow-up.
The presence and severity of MR were significantly associated with worse clinical functional status (Ross score). Figure 1 shows the relation between MR grade and Ross score (determined retrospectively from clinical assessment on the same day as echocardiography). Each increase of 1 MR severity grade was associated with a 0.64 ± 0.17 point increase in the Ross score (p = 0.001). The presence of MR was associated with higher hazard of death or transplantation at the initial echocardiographic exam (hazard ratio 2.4, p = 0.003) and at the last exam (hazard ratio 1.9, p = 0.009) ( Figure 2 ). In a multivariate survival regression model for which the EF, end-systolic volume, the E/E′ ratio, and functional class were entered into the model, MR remained an independent risk factor for increased hazard of death or heart transplantation.
