Mitral Valve Replacement in Infants Using a 15-mm Mechanical Valve.

BACKGROUND: The 15-mm mechanical valve was approved by the US Food and Drug Administration in March 2018. We review our experience in infants with this valve in the mitral position (MV), focusing on outcomes and timing to repeat MV replacement (MVR). METHODS: Between 2006 and 2017 7 patients underwent eight MVRs (one repeat) with a 15-mm mechanical valve. Retrospective chart review was performed to examine short- and long-term outcomes. RESULTS: There were no operative deaths. Mean follow-up was 5.8 ± 4.8 years (range, 0.72 to 11.1). Six patients underwent an MV operation 53 ± 39 days (range, 9 to 118) before MVR with the 15-mm valve. All patients were on mechanical ventilatory support at the time of operation. Mean age, body weight, and body surface area at time of 15-mm MVR were 0.5 ± 0.3 years (range, 0.2 to 0.9), 5.6 ± 0.8 kg (range, 4.8 to 6.6), and 0.29 ± 0.03 m2 (range, 0.27 to 0.32), respectively. Two patients required pacemaker implantation for atrioventricular block, both after their second MVR. Two patients are well at 16 and 24 months. Four patients underwent repeat MVR because of somatic growth and patient-prosthesis mismatch. Mean time to repeat MVR was 23 months (range, 6 to 40). There were two late deaths, one at 10 months unrelated to the valve in a child with a chromosomal abnormality. The other child had a congenital diaphragmatic hernia, early valve thrombosis, and died of multiple complications after a fourth MVR. CONCLUSIONS: The 15-mm mechanical valve was useful in treating MV disease in infants 2 to 12 months of age. This newly approved smallest available mechanical valve has a predicted mean time to replacement of 23 months in the mitral position.

Anomalous Aortic Origin of a Coronary Artery: Symptoms Do Not Correlate With Intramural Length or Ostial Diameter.

BACKGROUND: Anomalous aortic origin of a coronary artery (AAOCA) is a known cause of sudden death. Our hypothesis was that longer intramural length and smaller ostial diameter correlate with preoperative symptoms. If true, this would assist in the decision for surgical indications. We also assessed the accuracy of preoperative imaging to predict intramural length. METHODS: Retrospective analysis of patients who underwent AAOCA unroofing from 2006 to 2014. Patients had preoperative computed tomography angiography (CTA) or magnetic resonance imaging (MRI). Intramural length was measured. Intramural lengths and ostial diameters were also measured intraoperatively (operating room [OR]). Symptoms were noted. Intramural lengths and ostial diameters were compared between patients with and without preoperative symptoms. The accuracy of intramural length measured by CTA/MRI versus the length measured in the OR was assessed using a Bland-Altman analysis. RESULTS: Sixty-six patients underwent surgical repair of AAOCA. Fifty-two (79%) patients were symptomatic and 14 (21%) were asymptomatic. Mean age was 12.4 ± 4.0 years. There was no mortality. There was strong agreement between intramural length measured by CTA/MRI and measured in the OR. There was no significant difference in AAOCA intramural length in the symptomatic (8.6 ± 3.5 mm) and asymptomatic (8.9 ± 2.8 mm, P = .77) patients, which were measured both by CTA/MRI and intraoperatively (symptomatic 7.3 ± 2.5 mm, asymptomatic 6.9 ± 2.8 mm; P = .62). There was also no significant difference in AAOCA ostial diameters between groups (symptomatic = 1.9 ± 0.5 mm, asymptomatic = 1.6 ± 0.5 mm; P = .09). CONCLUSION: Preoperative CTA/MRI was very accurate in predicting the length of surgical unroofing. There was no demonstrable correlation between preoperative symptoms and intramural AAOCA length or AAOCA ostial diameter.

Discontinuous Pulmonary Artery.

We describe the diagnosis and surgical repair of a five-month-old infant with a congenital discontinuous right pulmonary artery. Initial echocardiogram failed to show the right pulmonary artery and revealed systemic left pulmonary artery pressure based on the tricuspid regurgitation jet. Computed tomographic angiography confirmed the diagnosis of discontinuous right pulmonary artery. The right pulmonary artery appeared essentially normal in size, and there were no significant aortopulmonary collateral arteries. Using cardiopulmonary bypass and aortic transection, we created an anastomosis between the right and the main pulmonary arteries augmented anteriorly by a pericardial patch. Postoperative lung perfusion scan demonstrated balanced pulmonary blood flow to the lungs. Pulmonary hypertension resolved over three weeks in the postoperative period, an expected outcome in this age-group.

Novel Modifications of a Ventricular Assist Device for Infants and Children.

BACKGROUND: A continuous-flow "adult" ventricular assist device (VAD) was modified to support infants and children waiting for heart transplantation. METHODS: A centrifugal VAD, designed to flow at 1.5 to 8 L/min, was used as a bridge to transplantation in pediatric patients. In smaller children and infants, a modified recirculation shunt permitted lower flow ranges. In hypoxic patients, an oxygenator was spliced into the circuit. RESULTS: From 2010 to 2015, the VAD was placed in 13 consecutive patients. Age ranged from 0.9 to 16 years (median, 7 years). Body surface area (BSA) ranged from 0.4 to 2.1 m(2) (median, 0.8 m(2)). Ten patients had a BSA less than 1.0 m(2). Four patients were receiving extracorporeal membrane oxygenation (ECMO) before VAD. Three patients had single-ventricle physiology. Five patients had a recirculation shunt and 3 underwent insertion of an oxygenator. Median time on the VAD was 20 days (range, 2-140 days). In patients with a recirculation shunt, mean patient flow was 1.5 L/min (mean flow/BSA, 2.7 L/min/m(2)), with mean total VAD flow of 3.4 L/min. Twelve patients underwent transplantation, and 1 patient underwent VAD explantation. All patients survived and were discharged at a median of 26 days (range, 17-83 days) after transplantation. Three patients experienced major bleeding events. There were 2 cerebrovascular accidents. VAD mortality dropped from 33% (3 of 9) during 2007 to 2010 to 0% (0 of 13) between 2011 and 2015 (p = 0.05). Wait-list mortality dropped from 10% (5 of 52) to 4% (4 of 91) for these periods (p = 0.29). CONCLUSIONS: The centrifugal VAD successfully supported pediatric patients awaiting heart transplantation. The modified recirculation shunt facilitated the successful support of patients in whom optimal flows were substantially lower than those recommended by the manufacturer. The design allows placement of an in-line oxygenator. Compared with pulsatile devices, use of this VAD was associated with a trend toward decreased mortality associated with VAD use.

Assessment of the anatomical variation to be found in the normal tricuspid valve.

The normal tricuspid valve is tri-leaflet, supported by tendinous cords, which are themselves supported by papillary muscles. There can be marked variation in the anatomy of the normal tricuspid valve, which must be understood to differentiate it from pathological malformations. The tricuspid valve of 100 normal heart specimens was examined. The three leaflets of the tricuspid valve, along with the papillary muscles supporting the zones of apposition were identified, and details of the anatomy recorded and analyzed. All three leaflets were identified in all 100 hearts. The septal leaflet had tendinous cord attachments in 93 specimens to the ventricular septum. The medial papillary muscle had a single head in the majority of specimens, supporting the zone of apposition with the antero-superior leaflet in 97 specimens. The anterior papillary muscle attached to the mid-portion of the antero-superior leaflet in 62 specimens, and supplied the zone of apposition between the antero-superior and inferior leaflets in 81 specimens. There were rough zone cord attachments to the antero-superior leaflets in all specimens. The inferior leaflet had basal cord attachments in 87 specimens, with attachments to multiple small muscular heads in 37 specimens. The inferior papillary muscle was well formed in only 58 specimens. Although certain features are relatively constant, multiple variations in the normal tricuspid valve have been identified. Knowledge of these normal variations is necessary in understanding the function of this complex valve apparatus, along with the potential for pathology.

Inferior and right-sided juxtaposition of the left atrial appendage with an unexpected type of inter-atrial communication.

We have re-investigated an unusual cardiac specimen with juxtaposition of the atrial appendages. The original description dates to 1962, when the autopsy was performed at the Children's Memorial Hospital in Chicago, now Ann & Robert H. Lurie Children's Hospital of Chicago. The heart was subsequently stored in the Farouk S. Idriss Cardiac Registry at the same institution. The specimen shows usual atrial arrangement, but with the morphologically left appendage juxtaposed in a rightward manner, passing behind the heart rather than through the transverse sinus so as to reach its location inferior to the morphologically right appendage. The heart also demonstrated an inter-atrial communication between the cavities of the juxtaposed left appendage and the morphologically right atrium. We provide a detailed description of the morphology, and provide images of this lesion, which to the best of our knowledge has not previously been described.

Aortic Implantation of Anomalous Origin of the Left Coronary Artery From the Pulmonary Artery: Long-Term Outcomes.

BACKGROUND: Since 1989 all patients with anomalous origin of the left coronary artery from the pulmonary artery at our institution have been treated with aortic implantation. The purpose of this review was to assess the late outcomes of these patients, especially regarding left ventricular (LV) function and mitral valve insufficiency. METHODS: Between 1989 and 2014, 36 patients had aortic implantation of anomalous origin of the left coronary artery from the pulmonary artery. Mean age at surgery was 2.5 ± 5.1 years (median, 0.5 years). Operative strategy included antegrade cold-blood cardioplegia, main pulmonary artery transection, aortic implantation with a large button of pulmonary artery, pulmonary reconstruction with fresh autologous pericardium, and prolonged postoperative inotropic and ventilator support. Mitral regurgitation and LV dysfunction were graded as 0 to 4 (0 = none, 1 = trivial, 1.5 = trivial-mild, 2 = mild, 2.5 = mild-moderate, 3 = moderate, 3.5 = moderate-severe, and 4 = severe). RESULTS: Mean mitral regurgitation grade preoperatively was 2.95 ± 0.95. Mean LV dysfunction grade was 3.14 ± 1.27. Mean cross-clamp and cardiopulmonary bypass times were 49.1 ± 18 minutes (median, 48.5 minutes) and 147.5 ± 45 minutes (median, 139 minutes), respectively. There was no operative or late mortality. Four patients had delayed sternal closure. Mean duration of ventilator support was 11 ± 6.6 days (median, 9 days). Two patients required 3 and 6 days of postoperative extracorporeal mechanical circulatory support. Mean length of stay was 25 ± 18 days (median, 19 days). No patient has required reoperation for supravalvar pulmonary stenosis, coronary stenosis, or mitral valve repair or replacement. Late echocardiographic follow-up shows a mean mitral regurgitation grade of 1.67 ± 1.05 and a mean LV dysfunction grade of 0.23 ± 0.68. CONCLUSIONS: Aortic implantation is our procedure of choice for patients with anomalous origin of the left coronary artery from the pulmonary artery. No patient required mitral valve repair or transplant. There was marked improvement of mitral regurgitation grade, return to essentially normal LV function, and no mortality during a 25-year period.

Modified single patch: are we still worried about subaortic stenosis?

BACKGROUND: When the modified single-patch technique for atrioventricular septal defect (AVSD) repair was introduced by Dr Benson Wilcox, there was concern that these patients might be at risk for late subaortic stenosis and left ventricular outflow tract obstruction (LVOTO). This review evaluated our modified single-patch population for LVOTO in the postoperative period. METHODS: Between January 2000 and 2013, 77 infants underwent AVSD repair with a modified single-patch technique. Median age was 4.2 months, and median weight was 5 kg. Eight patients had a prior repair of coarctation of the aorta via left thoracotomy in the newborn period. RESULTS: The median hospital stay was 10 days. No patient required a pacemaker. The mean and median follow-up times were 4.6 and 3.7 years, respectively. Only 2 patients (2.5%) required reoperation for LVOTO; both had prior repair of coarctation of the aorta (2 of 8 vs 0 of 69, p = 0.01). A discrete fibrous subaortic membrane developed in the first patient that required resection at 3 and 7 years after repair. The other patient had LVOTO from accessory chordae of the left atrioventricular valve and required mitral valve replacement 5 months after repair. One early death occurred at 4 months postoperatively due to liver failure related to hyperalimentation. CONCLUSIONS: At intermediate term follow-up, LVOTO does not appear to be a significant postoperative issue after modified single-patch repair of AVSD. Coarctation of the aorta was the most significant predictor of late LVOTO after repair of AVSD with the modified single-patch technique.

Exercises in anatomy: holes between the ventricles.

Holes between the ventricles are the commonest congenital cardiac malformations. As yet, however, there is no consensus as to how they can best be described and categorized. In this, our third exercise in cardiac anatomy, we address the issue of classification of ventricular septal defects. We begin our demonstration by analysing the normal heart. We show that the larger part of the ventricular septum is made up of its muscular component. The membranous part accounts for only a small portion, which is located centrally within the cardiac base. This small membranous part forms a boundary between the right-sided chambers and the aortic root. Holes at this site, therefore, which account for the commonest defects closed surgically, will open centrally in the cardiac base, being located postero-inferiorly relative to the supraventricular crest. We then show that the larger part of the crest itself is a free-standing muscular sleeve, which lifts the leaflets of the pulmonary valve away from the cardiac base. Only a very small part of the muscle forming the right ventricular outlet is located in the septal position. Turning our attention to malformed hearts, we show how holes between the ventricles can open centrally at the cardiac base, open to the inlet or outlet of the right ventricle or open within the substance of the apical muscular septum. We demonstrate, however, that description of such geographical location of the defects does not paint the full picture, since lesions with markedly different phenotypic features can open in comparable geographic locations. We illustrate how it is the phenotypic features, as seen from the right ventricle, which convey the crucial information for the surgeon with regard to the location of the atrioventricular conduction axis, using hearts with holes opening to the inlet of the right ventricle with muscular as opposed to partially fibrous borders to emphasize this point. We continue by showing how holes with different phenotypes can also open to the outlet of the right ventricle, the key feature in this regard being malalignment between the apical muscular septum relative to the muscular outlet septum or its fibrous remnant. Malalignment can also be found between the apical ventricular septum and the atrial septum, this being shown in a defect opening to the inlet of the right ventricle. We conclude by emphasizing that, so as to bring together all the information of surgical significance, it is necessary to take note of the geographical location of holes between the ventricles, their phenotypic features and the presence or absence of malalignment between the septal components.

Exercises in anatomy: cardiac isomerism.

It is well recognized that the patients with the most complex cardiac malformations are those with so-called visceral heterotaxy. At present, it remains a fact that most investigators segregate these patients on the basis of their splenic anatomy, describing syndromes of so-called asplenia and polysplenia. It has also been known for quite some time, nonetheless, that the morphology of the tracheobronchial tree is usually isomeric in the setting of heterotaxy. And it has been shown that the isomerism found in terms of bronchial arrangement correlates in a better fashion with the cardiac anatomy than does the presence of multiple spleens, or the absence of any splenic tissue. In this exercise in anatomy, we use hearts from the Idriss archive of Lurie Children's Hospital in Chicago to demonstrate the isomeric features found in the hearts obtained from patients known to have had heterotaxy. We first demonstrate the normal arrangements, showing how it is the extent of the pectinate muscles in the atrial appendages relative to the atrioventricular junctions that distinguishes between morphologically right and left atrial chambers. We also show the asymmetry of the normal bronchial tree, and the relationships of the first bronchial branches to the pulmonary arteries supplying the lower lobes of the lungs. We then demonstrate that diagnosis of multiple spleens requires the finding of splenic tissue on either side of the dorsal mesogastrium. Turning to hearts obtained from patients with heterotaxy, we illustrate isomeric right and left atrial appendages. We emphasize that it is only the appendages that are universally isomeric, but point out that other features support the notion of cardiac isomerism. We then show that description also requires a full account of veno-atrial connections, since these can seemingly be mirror-imaged when the arrangement within the heart is one of isomerism of the atrial appendages. We show how failure to recognize the presence of such isomeric appendages can lead to spurious diagnoses of discordant atrioventricular connections. Overall, we show that the finding of isomeric atrial appendages, guided by the finding of bronchial isomerism, is but the prelude to full segmental sequential analysis.

Modified TandemHeart ventricular assist device for infant and pediatric circulatory support.

PURPOSE: The development of pediatric ventricular assist device (VAD) circuits with lower flow ranges for infants and small children is ongoing. We present our results with modifying a readily available adult VAD to support the pediatric population. DESCRIPTION: The TandemHeart VAD (CardiacAssist, Pittsburgh, PA) circuit was modified to include a variable restrictive recirculation shunt to permit lower flow ranges in small pediatric patients. EVALUATION: Initial benchtop flow rates and pressures were studied. Hemolysis trials were performed using whole bovine blood to compare plasma-free hemoglobin levels between modified and unmodified VAD circuits. The modified VAD was surgically implanted in 7 piglets (6 to 14 kg) and which supported them for 4 hours. Levels of hemolysis did not increase and full hemodynamic support was achieved. The modified TandemHeart VAD with a recirculation shunt was subsequently implanted in 2 pediatric patients who were bridged to transplant successfully. CONCLUSIONS: Because of its simplicity, availability, low prime volume, greater patient flow range, and lower cost, the modified TandemHeart VAD with a recirculation shunt should be considered as an alternative to extracorporeal membrane oxygenation and other pulsatile VADs in children.