The study enrolled 26 patients who underwent 37 total Melody valve implants into the atrioventricular valve position. The valve was placed in the systemic atrioventricular (mitral or systemic tricuspid) valve position in 31 cases (83.8%). There were 3 total patients (11.6%) with single ventricle physiology and the remainder had biventricular circulation. Baseline patient characteristics are outline in Table 1.
Table 1
Baseline Patient Demographics (n = 26)
Characteristics | Value |
Male | 13 (50%) |
Genetic syndrome | 15 (57.7%) |
Trisomy 21 | 6 |
10q11 deletion | 1 |
KBG syndrome | 1 |
Hajdu-Cheney syndrome | 1 |
15q11.2 deletion | 1 |
Trisomy 18 | 1 |
8p23.1 microdeletion | 1 |
Primary cardiac diagnosis | |
Complete atrioventricular septal defect | 11 (42.3%) |
Hypoplastic left heart syndrome | 3 (11.5%) |
Pulmonary atresia with intact ventricular septum | 3 (11.5%) |
Double outlet right ventricle, mitral atresia | 2 (7.7%) |
Shone’s complex | 2 (7.7%) |
Partial atrioventricular septal defect | 2 (7.7%) |
Ventricular septal defect with coarctation | 1 (3.8%) |
Congenital dysplastic tricuspid valve | 1 (3.8%) |
Congenital dysplastic mitral valve | 1 (3.8%) |
There were 50% male and 50% female patients. Over half of the patients in the study had a genetic diagnosis with Trisomy 21 being the most common. The most common cardiac diagnosis that required Melody valve placement in the atrioventricular valve position was complete atrioventricular septal defect (42.3%). Indications for valve replacement were valve regurgitation in 13 (35.1%), valve stenosis in 12 (32.4%) and mixed in 12 (32.4%). Function of the subvalvar ventricle was normal in 23 (62.2%), mildly depressed in 7 (18.9%), moderately depressed in 4 (10.8%) and severely depressed in 3 (8.1%). In 21 of the 37 valve replacements (56.8%), the patient was hospitalized prior to the intervention. All but 1 patient (96.2%) had undergone prior surgical or catheter-based intervention before the Melody valve was placed.
Procedural characteristics are detailed below in Table 2. Median age at procedure was 17 mos (IQR 4–33) and weight was 8.5 kg (IQR 6.3–12.9). Median balloon size for valve implant was 20 mm (IQR 18–22). An expected valve annulus diameter (Z score of 0) was calculated for all patients and this was utilized to compare to the balloon size chosen for patients. There were 4 total cases (10.8%) that had procedural related complications. Two patients had cardiac arrest requiring ECMO immediately following the operation, one patient had mediastinitis and one had complete heart block. Median length of stay following procedure was 13 days (IQR 5-52.5).
Table 2
Procedural Characteristics (n = 37)
Variable | Mean | SD | Median | IQR |
Age (months) | 24.2 | 26.2 | 17.0 | 4.0–33.0 |
Weight (kg) | 10.3 | 6.5 | 8.5 | 6.3–12.9 |
BSA (m2) | 0.5 | 0.2 | 0.4 | 0.3–0.6 |
Bypass Time (min) | 127.5 | 92.7 | 89.0 | 75.0-167.0 |
Predicted Z0 valve diameter (mm) | 15.0 | 3.2 | 14.8 | 12.7–22.8 |
Balloon size (mm) | 19.2 | 2.5 | 20.0 | 18.0–22.0 |
Balloon size / Z0 valve diameter | 1.3 | 0.2 | 1.3 | 1.2–1.4 |
Balloon size / BSA (mm/m2) | 49.1 | 17.4 | 45.5 | 36.4–58.4 |
LOS (days) | 38.9 | 61.5 | 13.0 | 5.0-52.5 |
Patients were followed for a median length of 30 months following valve implant (IQR 3-38.5). Total available patient years of follow-up is 84.73. Median freedom from re-intervention was 31 months; 19% were free from re-intervention at 60 months as shown below in Fig. 2.
Re-intervention occurred in 21 of the 37 valve implants (56.8%) with all but one of these being a repeat surgical intervention. The one other intervention was a catheter-based balloon dilation of the Melody prosthesis. Of the 21 cases requiring re-intervention, the mechanism of valve failure was recorded as mixed stenosis and regurgitation in 11 (52.4%), stenosis in 7 (33.3%), endocarditis in 1 (4.8%), thrombus in 1 (4.8%) and LVOT obstruction in 1 (4.8%). In no cases was pure regurgitation the reason for valve failure.
Of the 26 total patients, 10 patients (38%) have had no further interventions since the Melody was placed. Six patients had a repeat Melody valve placed when the first one failed, one patient had three Melody valves placed and one patient had four total Melody valves (three in the mitral position and one in the tricuspid position).
There were seven total deaths with a Melody valve in place (27%) and 1 death after a Melody valve was explanted leaving 18 living patients at last follow-up (69%). The reasons for death with Melody valve in place included septic shock, necrotic bowel, acute respiratory failure, coagulopathy on ECMO, cardiopulmonary arrest, hemorrhagic shock and cardiac failure on LVAD with associated ischemic stroke. All of the deaths with Melody valve in place were within 6 months after Melody valve placement.
Time to valve failure based on different variables is shown below in Table 3. Age < 12 months (p = 0.003), weight < 10 kg (p = 0.025) and BSA < 0.4 m2 (p = 0.017) all discriminate subjects at high risk (small, young) from those at lower risk (large, older) of early valve failure, however these are all highly intercorrelated. Single ventricle physiology (p = 0.027) is also a risk for early valve failure but this is essentially a phenomenon seen only in those patients less than 12 months of age.
Implantation at a smaller balloon size (in absolute diameter) correlates with higher risk of early valve failure (p = 0.017), but this is probably also a manifestation of patient size. Intermediate balloon size (1.20-1.35x) relative to valve annulus Z score of 0 is associated with the best outcome, with more early valve failure with both smaller and larger balloons relative to Z0 (p = 0.038).
Table 3
Time to Valve Failure Based on Different Variables
Variable | Mean + SEE time to failure (mos) | |
With feature | Without feature | p |
Gender (male) | 57.7 ± 11.7 | 27.0 ± 4.2 | 0.074 |
Single ventricle physiology | 8.3 ± 4.7 | 48.4 ± 7.1 | 0.027 |
Genetic syndrome | 35.7 ± 3.9 | 35.8 ± 9.9 | 0.167 |
Systemic valve position | 39.3 ± 6.9 | 48 | 0.086 |
Inpatient prior to procedure | 29.0 ± 4.9 | 52.1 ± 9.9 | 0.159 |
Indication for implant (stenosis) | 56.2 ± 11.6 | | 0.374 |
Indication for implant (regurgitation) | 26.3 ± 3.7 |
Indication for implant (mixed) | 26.7 ± 6.7 |
Subvalvar ventricular dysfunction (none) | 51.6 ± 8.5 | | 0.129 |
Subvalvar ventricular dysfunction (mild) | 23.2 ± 6.1 |
Subvalvar ventricular dysfunction (moderate/severe) | 29.6 ± 12.8 |
Age < 12 mos | 22.3 ± 5.0 | 61.8 ± 9.9 | 0.003 |
BSA < 0.4 m2 | 25.5 ± 4.5 | 65.1 ± 11.4 | 0.017 |
Weight < 10.0 kg | 27.3 ± 4.1 | 66.9 ± 12.6 | 0.025 |
Balloon size ≤ 16 mm | 24.5 ± 8.9 | | 0.017 |
Balloon size 18 or 20 | 26.7 ± 4.5 |
Balloon size ≥ 22 | 72.6 ± 12.6 |
Balloon size / Z0 ≤ 1.20 | 23.8 ± 5.4 | | 0.038 |
Balloon size / Z0 1.20–1.35 | 64.2 ± 11.1 |
Balloon size / Z0 ≥ 1.35 | 23.4 ± 5.6 |
Freedom from valve failure is lower in patients < 12 months of age (p = 0.03) as seen below in Fig. 3. Patients less than 12 months of age had a mean time to valve failure of 22.3 months compared to 61.8 months in patients older than 12 months of age.
Those patients less than 12 months of age, as a higher risk group at baseline, were then analyzed and different variables were assessed for time to valve failure. These variables are shown below in Table 4.
Table 4
Time to Valve Failure Based on Different Variables in Patients < 12 months of age
Variable | Mean + SEE time to failure (mos) | |
With feature | Without feature | p |
Gender (male) | 25.4 ± 8.1 | 20.2 ± 6.6 | 0.895 |
Single V Physiology | 5.6 ± 4.8 | 28.8 ± 5.6 | 0.004 |
Syndrome | 23.3 ± 6.7 | 20.9 ± 7.5 | 0.991 |
Valve Position (systemic) | 20.2 ± 5.0 | 48 | 0.507 |
Inpatient prior to procedure | 21.4 ± 5.9 | 20.8 ± 8.2 | 0.383 |
Indication for implant (stenosis) | 8.4 ± 8.4 | | 0.533 |
Indication for implant (regurgitation) | 22.1 ± 5.7 |
Indication for implant (mixed) | 13.8 ± 11.5 |
Subvalvar ventricular dysfunction (none) | 22.6 ± 6.1 | | 0.476 |
Subvalvar ventricular dysfunction (mild) | 11.7 ± 11.2 |
Subvalvar ventricular dysfunction (moderate/severe) | 34.5 ± 15.5 |
Balloon size ≤ 16 mm | 28.6 ± 9.4 | | 0.252 |
Balloon size 18 or 20 | 18.5 ± 5.8 |
Balloon size / Z0 ≤ 1.4 | 18.0 ± 6.6 | | 0.311 |
Balloon size / Z0 ≥ 1.4 | 27.5 ± 7.7 |
The only significant finding in patients < 12 months is single ventricle physiology with a mean time to valve failure of only 5.6 months as compared to 28.8 months in patients with biventricular physiology (p = 0.004). This is also shown in a Kaplan-Meier curve below in Fig. 4. No other feature, including balloon size and normalized balloon size has a significant association with early valve failure.
Patients then greater than or equal to 12 months of age were analyzed as a separate group with a baseline lower risk of early valve failure. These variables are detailed below in Table 5.
Table 5
Time to Valve Failure Based on Different Variables in Patients >/= 12 months of age
Variable | Mean + SEE time to failure (mos) | |
With feature | Without feature | p |
Gender (male) | 83.8 ± 15.4 | 31.5 ± 3.9 | 0.029 |
Single V Physiology | undefined | 59.8 ± 10.0 | 0.436 |
Syndrome | 37.2 ± 1.1 | 46.4 ± 19.1 | 0.128 |
Systemic valve Position (systemic) | 54.8 ± 10.7 | undefined | 0.160 |
Inpatient prior to procedure | 57.4 + 11.5 | 30.0 | 0.564 |
Indication for implant (stenosis) | 75.0 ± 14.9 | | 0.464 |
Indication for implant (regurgitation) | 31.5 ± 0.6 |
Indication for implant (mixed) | 33.0 ± 5.5 |
Subvalvar ventricular dysfunction (any) | 26.3 + 5.0 | 70.8 ± 10.9 | 0.006 |
Balloon size < 22 | 30.8 ± 4.0 | | 0.021 |
Balloon size ≥ 22 | 78.6 ± 12.8 | | |
Balloon size / Z0 ≤ 1.20 | 29.8 ± 5.0 | | 0.562 |
Balloon size / Z0 ≥ 1.20 | 64.4 ± 12.8 |
Among older patients, female gender (p = 0.029, Fig. 5), subvalvar ventricular dysfunction (p = 0.006, Fig. 5, and absolute balloon size less than 22 mm (p = 0.021, Fig. 5) are all risk factors for early valve failure. This dataset is insufficient to establish whether these factors act independently of one another.