Fluoroscopy-guided Umbilical Venous Catheter Placement in Infants with Congenital Heart Disease.

OBJECTIVE: The objective of this study was to (1) describe the technical aspects of fluoroscopy-guided umbilical venous catheter placement (FGUVCP); and (2) determine the procedural success rate, factors contributing to procedural failure, and risks of the procedure. BACKGROUND: Umbilical venous catheters are advantageous compared with femoral venous access, but can be difficult to place at the bedside. MATERIALS AND METHODS: This was a retrospective chart review from a single tertiary care referral institution. RESULTS: FGUVCP was successful in 138 of 180 patients (76.7%) over a seven-year period. Patients in whom FGUVCP was successful were younger at the time of procedure compared with patients in whom FGUVCP was unsuccessful (median 18.2 vs. 22.2 hours, P = .03). The optimal age cutoff to predict FGUVCP success was 20 hours with a high positive predictive value (82.4%) but low negative predictive value (32.5%). No other variables were associated with procedural failure, though functional univentricular heart and older gestational age trended toward statistical significance. Median radiation time, contrast exposure, and blood loss were 3.2 minutes, 1 mL, and 1 mL, respectively. A total of 10 complications in 10 patients were associated with FGUVCP. CONCLUSIONS: FGUVCP is a safe and highly successful way to obtain central venous access in neonates with congenital heart disease. Older age at the time of procedure is associated with procedural failure, but utilization of an age cutoff may not be clinically useful.

Long-term follow up of secundum atrial septal defect closure with the amplatzer septal occluder.

BACKGROUND: During the past 15 years, closure of a secundum atrial septal defect (ASD) has moved from a surgical to a percutaneous transcatheter approach. Few long-term studies of the efficacy and safety of closure of an ASD by an Amplatzer septal occluder (ASO) exist. METHODS: To examine the long-term results of secundum ASD closure using the ASO, data on 94 patients who underwent secundum ASD closure with the ASO between 1998 and 2002 were available and reviewed. Data regarding residual shunt, chest pain, palpitations, arrhythmias, headaches, transient ischemic attacks, cerebrovascular accidents, and mortality were collected. RESULTS: Seven (7.4%) subjects had residual shunts immediately following ASO placement. During follow-up, 4 residual shunts closed for a complete closure rate of 97%. Eighteen (19%) patients reported chest pain during the follow-up period. Twenty-three patients (24%) reported palpitations during the follow up period, 7 were documented arrhythmias, including supraventricular tachycardia, atrial fibrillation, and premature ventricular beats. Migraine headaches were new-onset in 4 patients. Migraine cessation occurred in 2 patients after secundum ASD closure. One child died from a cerebral vascular event 18 months following device placement. Only 1 patient developed mild aortic insufficiency. CONCLUSION: These data indicate that for up to 120 months of patient follow-up, the ASO continues to be a safe device. Residual shunts and arrhythmias have low incidence post-ASO placement. Given the mortality in one high-risk patient, further investigation into anti-platelet therapy after device placement is warranted.

Use of mathematic modeling to compare and predict hemodynamic effects of the modified Blalock-Taussig and right ventricle-pulmonary artery shunts for hypoplastic left heart syndrome.

OBJECTIVE: Stage one reconstruction (Norwood operation) for hypoplastic left heart syndrome can be performed with either a modified Blalock-Taussig shunt or a right ventricle-pulmonary artery shunt. Both methods have certain inherent characteristics. It is postulated that mathematic modeling could help elucidate these differences. METHODS: Three-dimensional computer models of the Blalock-Taussig shunt and right ventricle-pulmonary artery shunt modifications of the Norwood operation were developed by using the finite volume method. Conduits of 3, 3.5, and 4 mm were used in the Blalock-Taussig shunt model, whereas conduits of 4, 5, and 6 mm were used in the right ventricle-pulmonary artery shunt model. The hydraulic nets (lumped resistances, compliances, inertances, and elastances) were identical in the 2 models. A multiscale approach was adopted to couple the 3-dimensional models with the circulation net. Computer simulations were compared with postoperative catheterization data. RESULTS: Good correlation was found between predicted and observed data. For the right ventricle-pulmonary artery shunt modification, there was higher aortic diastolic pressure, decreased pulmonary artery pressure, lower Qp/Qs ratio, and higher coronary perfusion pressure. Mathematic modeling predicted minimal regurgitant flow in the right ventricle-pulmonary artery shunt model, which correlated with postoperative Doppler measurements. The right ventricle-pulmonary artery shunt demonstrated lower stroke work and a higher mechanical efficiency (stroke work/total mechanical energy). CONCLUSIONS: The close correlation between predicted and observed data supports the use of mathematic modeling in the design and assessment of surgical procedures. The potentially damaging effects of a systemic ventriculotomy in the right ventricle-pulmonary artery shunt modification of the Norwood operation have not been analyzed.

Trans-hepatic cannulation: a novel approach for placement of a pediatric percutaneous ventricular assist device.

Pediatric ventricular assist device (VAD) use is increasing. The TandemHeart PTVA system is a percutaneous VAD used in adults. Pediatric use is limited by cannula size. Trans-hepatic left atrial (LA) cannulation offers the possibility of shorter, larger-diameter cannulae, but specific requirements for cannula size and flow characteristics need to be developed. We evaluated the feasibility of trans-hepatic cannulation based on patient measurements and bench-top modeling. Two hundred patients admitted to the University of Michigan Pediatric CICU were evaluated. Patients over 14 years and dextrocardiac patients were excluded. The distance from the skin between the 10th and 11th ribs at the mid-axillary line to the left-heart border was measured on x-ray. LA size was measured on echocardiogram. From these measurements, prototype cannulae were fabricated. Benchtop testing of developed cannulae was performed with a pump speed of 8500 RPM and fluid viscosity of 4.0 cP at physiologic atrial and arterial pressures. Inlet pressure and deliverable flows were measured. The proposed design delivered sufficient flow with negative pressures low enough to avoid hemolysis. Trans-hepatic LA cannulation for percutaneous VAD placement is feasible. The shorter distance to the LA and larger hepatic vein size allows design of pediatric-appropriate cannulae with adequate flow rates.

Catheter-based decompression of the left atrium in patients with hypoplastic left heart syndrome and restrictive atrial septum is safe and effective.

Infants with hypoplastic left heart syndrome (HLHS) and restrictive or intact atrial septum (rAS) present with cyanosis, pulmonary edema, and are critically ill. A previous report from our institution on emergent Norwood for HLHS with rAS showed 10% survival. We hypothesized that transcatheter left atrial (LA) decompression in HLHS with rAS would safely and effectively relieve LA hypertension, improve oxygenation, and improve Norwood survival. Between 1996 and 2004, 30 patients with HLHS and rAS underwent cardiac catheterization for pre-Norwood intervention. Twenty-eight atrial septostomies were performed: 23 static balloon dilations, 4 Rashkind septostomies, and 1 intra-atrial stent. Two procedures were aborted due to perforation (n = 1) or inability to enter the LA (n = 1). Eight total patients required surgical septectomy, for a failure rate of 27%. There were no catheter-related mortalities, although two patients died within 36 hr of the procedure after surgical septectomy. Major complications occurred in three patients (10%)--atrial perforations requiring intervention. Mean atrial septal defect gradient fell from 16.7 +/- 4.9 to 6.3 +/- 3.4 mm Hg (P < 0.001; n = 18). Mean LA pressure dropped from 21.8 +/- 5.5 to 13.1 +/- 6.5 mm Hg (P < 0.001; n = 16). Mean PaO(2) rose from 29.5 +/- 9.1 to 36.5 +/- 5.1 torr (P < 0.001; n = 23). Seventeen of 30 patients (57%) survived to discharge from Norwood. Thirteen have undergone hemi-Fontan and nine Fontan. Sixteen of 22 successful decompressions (73%) survived to discharge. Transcatheter decompression of the LA for patients with HLHS and rAS can be performed safely, reduces the transatrial gradient, and improves oxygenation. Catheter intervention improves survival compared to historical controls undergoing emergent Norwood.

Transcatheter occlusion of aortopulmonary shunts during single-ventricle surgical palliation.

Development of aortopulmonary collaterals during the course of surgical palliation for single-ventricular anatomy has been linked to adverse outcomes following Fontan palliation. We investigated the hemodynamic significance of aortopulmonary collaterals during presurgical cardiac catheterization of patients with single-ventricle surgically palliated anatomy. Thermal indicator dilution studies were performed to determine degree of shunt. A total of 52 patients were studied and the data were analyzed. Measurements by thermal indicator dilution correlated significantly with qualitative angiographic grading of aortopulmonary collaterals. However, the hemodynamic significance of these aortopulmonary collaterals, as measured by thermal indicator dilution, did not correlate with postoperative outcome variables. This study demonstrated that thermal indicator dilution measurement holds promise for hemodynamically quantifying the significance of aortopulmonary shunts in surgically palliated single-ventricular patients. However, the degree of recirculation from aortopulmonary collaterals does not correlate with outcome after Fontan surgery in this patient cohort.

Comparison of simultaneous invasive and noninvasive measurements of pressure gradients in congenital aortic valve stenosis.

PURPOSE: Congenital aortic valve stenosis is a common problem in pediatric cardiology. The catheter peak to peak systolic gradient is the accepted standard used for prognosis and intervention, but noninvasive correlation in pediatric patients is frequently associated with underestimation or overestimation of this gradient. The purpose of this study was to compare different noninvasive measurements with simultaneous catheter gradients to identify which best predicts the catheter peak to peak gradient. METHODS: Twenty-five simultaneous Doppler and catheter measurements of aortic stenosis gradient were performed in 14 children (all 14 before valvuloplasty and 11 after valvuloplasty). Noninvasive estimates of pressure gradient were compared with catheter measurements with linear regression and Bland-Altman analysis. RESULTS: The Doppler peak instantaneous pressure gradient overestimated the catheter peak to peak gradient but correlated well with the catheter peak instantaneous gradient. The Doppler mean systolic gradient correlated well with the catheter peak to peak gradient at low gradients and underestimated higher catheter gradients but agreed well at all levels with the catheter mean gradient. The modification of a catheter-derived correlation equation produced good correlation with the catheter peak to peak gradient (slope, 1.14; intercept, -1.8; R, 0.92), as did the use of estimated pressure recovery (slope, 1.04; intercept, 5.0; R, 0.94), calculated from a defined fluid mechanic equation. CONCLUSION: The catheter peak to peak gradient can be accurately estimated noninvasively using estimated pressure recovery or correlation equations incorporating Doppler measurements.

Influence of Echocardiographic Guidance on Positioning of the Buttoned Occluder for Transcatheter Closure of Atrial Septal Defects.

Ideal position of the buttoned device for occlusion of atrial septal defects was achieved more often with transesophageal than with transthoracic echocardiographic guidance (10/11 vs 7/23; P = 0.001). Patients with ideal device position were less likely to have residual shunts, device unbuttoning, or atrioventricular valve regurgitation (2/17 vs 11/17; P = 0.002). We therefore recommend the use of transesophageal echocardiography to guide implantation of the buttoned device. (ECHOCARDIOGRAPHY, Volume 13, March 1996)