Toward designing the optimal total cavopulmonary connection: an in vitro study

doi:10.1016/S0003-4975(99)00560-3

The Annals of Thoracic Surgery

Volume 68, Issue 4, October 1999, Pages 1384-1390

https://doi.org/10.1016/S0003-4975(99)00560-3 Get rights and content

Abstract

Background. Understanding the total cavopulmonary connection (TCPC) hemodynamics may lead to improved surgical procedures which result in a more efficient modified circulation. Reduced energy loss will translate to less work for the single ventricle and although univentricular physiology is complex, this improvement could contribute to improved postoperative outcomes. Therefore to conserve energy, one surgical goal is optimization of the TCPC geometry. In line with this goal, this study investigated whether addition of caval curvature or flaring at the connection conserves energy.

Methods. TCPC models were made varying the curvature of the caval inlet or by flaring the anastomosis. Steady flow pressure measurements were made to calculate the power loss attributed to each connection design over a range of pulmonary flow splits (70:30 to 30:70). Particle flow visualization was performed for each design and was qualitatively compared to the power losses.

Results. Results indicate that curving the cavae toward one pulmonary artery is advantageous only when the flow rate from that cavae matches the flow to the pulmonary artery. Under other pulmonary flow split conditions, the losses in the curved models are significant. In contrast, fully flaring the anastomosis reduced losses over the range of pulmonary flow splits. Power losses were 56% greater for the curving as compared to flaring. Fully flaring without caval offset reduced losses 45% when compared to previous models without flaring. If flaring on all sides was implemented with caval offset, power losses reduced 68% compared to the same nonflared model.

Conclusions. The results indicate that preferentially curving the cavae is only optimal under specific pulmonary flow conditions and may not be efficient in all clinical cases. Flaring of the anastomosis has great potential to conserve energy and should be considered in future TCPC procedures.

Section snippets

Material and methods

Glass models of a TCPC were custom-crafted, varying the geometry of the cavopulmonary anastomosis. A total of 9 models were studied to investigate the effects of curvature or flaring at the TCPC on the efficiency of the new geometry.

Curved models

Figure 3 shows the results for the curved models. The same trend can be seen in all cases. At a flow split of 70% to the RPA (right end of the graph), the power losses for all models were in the range of 4.5 to 6.5 mW. The minimum power losses were measured when the inferior vena cava and right pulmonary artery flow rates were equal (60% of the flow to the RPA). As this flow split to the RPA decreased, an increase in power loss was observed. In other words, as more flow was forced toward the

Curved models

The curved models were made with the inferior vena cava curved toward the right pulmonary artery and the superior cava toward the left. This explains why the power losses were lowest when the percentage of flow going to the RPA matched that entering from the IVC (60% of the total caval flow). In this case, all of the flow from the IVC follows the natural curve of the vessel toward the RPA, and all of the flow from the SVC follows the natural curve of the vessel toward the LPA. At this flow

Acknowledgements

We acknowledge the clinical expertise and advice of the cardiologists and cardiac surgeons at Egleston Children’s Hospital, Atlanta, GA and the skillful work of Jerry A. Cloninger in blowing the glass models for this study. We also acknowledge the financial support of the American Heart Association, Georgia Affiliate and the National Institutes of Health grant number R01 HL-52799.

References (12)

R.F.A Lins et al.
Orthoterminal correction of congenital heart diseasedouble cava-pulmonary anastomosis
J Thorac Cardiovasc Surg
(1982)
S Sharma et al.
In vitro flow experiments for determination of optimal geometry of total cavopulmonary connection for surgical repair of children with functional single ventricle
J Am Coll Cardiol
(1996)
M.R De Leval et al.
Use of computational fluid dynamics in the design of surgical proceduresapplication to the study of competitive flows in cavopulmonary connections
J Thorac Cardiovasc Surg
(1996)
G Gross et al.
Maturational and hemodynamic factors predictive of increased cyanosis after bidirectional cavopulmonary anastomosis
Am J Cardiol
(1994)
S.A Rebergen et al.
Postoperative pulmonary flow dynamics after Fontan surgeryassessment with nuclear magnetic resonance velocity mapping
J Am Coll Cardiol
(1993)
F Fontan et al.
Surgical repair of tricuspid atresia
Thorax
(1971)

There are more references available in the full text version of this article.

Cited by (108)

The impact of compliance on Stage 2 uni-ventricular heart circulation: An experimental assessment of the Bidirectional Glenn
2020, Medical Engineering and Physics
Citation Excerpt :
It has been well established that minimizing energy losses through these connections is essential in achieving an efficient circulation. A number of studies have utilized in-vitro and computational modelling to evaluate the hemodynamics of these operations in an effort to construct better designs [7–13]. In-vitro and in-silico studies have previously reported energy losses across the three surgical stages of Norwood [5], Glenn [1] and Fontan procedures [6–11,15–21].
The Bidirectional Glenn (BDG) or cavopulmonary connection is typically undertaken to volume unload the single ventricle in an effort to preserve ventricular and atrioventricular valve function. The geometry of this surgical palliation has been shown to influence the fluid energy loss as well as the distribution of flow that enters through the superior vena cava. In-vitro and in-silico studies to date have been performed on rigid wall models, while this investigation looks at the impact of flexible thin walled models versus rigid walls. Rigid and compliant models of two patient-specific Glenn geometries were fabricated and tested under various flow conditions, within a biosimulator capable of replicating patient specific flow conditions. It was found that the compliant models exhibit greater levels of energy loss compared to the rigid models. Along with these findings greater levels of turbulence was found in both compliant models compared to their rigid counterparts under ultrasound examinations. This shows that vessel compliance has a significant impact on the hemodynamics within hypoplastic left heart syndrome.
Evaluation of pulmonary artery stenosis in congenital heart disease patients using functional diagnostic parameters: An in vitro study
2018, Journal of Biomechanics
Congenital pulmonary artery (PA) stenosis is often associated with abnormal PA hemodynamics including increased pressure drop (Δp) and reduced asymmetric flow (Q), which may result in right ventricular dysfunction. We propose functional diagnostic parameters, pressure drop coefficient (CDP), energy loss (E_loss), and normalized energy loss ( ${\bar{E}}_{loss}$ ) to characterize pulmonary hemodynamics, and evaluate their efficacy in delineating stenosis severity using in vitro experiments. Subject-specific test sections including the main PA (MPA) bifurcating into left and right PAs (LPA, RPA) with a discrete LPA stenosis were manufactured from cross-sectional imaging and 3D printing. Three clinically-relevant stenosis severities, 90% area stenosis (AS), 80% AS, and 70% AS, were evaluated at different cardiac outputs (COs). A benchtop flow loop simulating pulmonary hemodynamics was used to measure Q and Δp within the test sections. The experimental Δp-Q characteristics along with clinical data were used to obtain pathophysiologic conditions and compute the diagnostic parameters. The pathophysiologic Q_LPA decreased as the stenosis severity increased at a fixed CO. CDP_LPA, E_loss,LPA (absolute), and ${\bar{E}}_{loss, L P A}$ (absolute) increased with an increase in LPA stenosis severity at a fixed CO. Importantly, CDP_LPA and ${\bar{E}}_{loss, L P A}$ had reduced variability with CO, and distinct values for each LPA stenosis severity. Under variable CO, a) CDP_LPA values were 14.5–21.0 (70% AS), 60.7– 2.2 (80% AS), $\geq$ 261.6 (90% AS), and b) ${\bar{E}}_{loss, L P A}$ values (in mJ per $Q_{LPA}$ ) were −501.9 to −1023.8 (70% AS), −1247.6 to −1773.0 (80% AS), −1934.5 (90% AS). Hence, CDP_LPA and ${\bar{E}}_{loss, L P A}$ are expected to assess the true functional severity of PA stenosis.
Role of Computational Modelling in Planning and Executing Interventional Procedures for Congenital Heart Disease
2017, Canadian Journal of Cardiology
Increasingly, computational modelling and numerical simulations are used to help plan complex surgical and interventional cardiovascular procedures in children and young adults with congenital heart disease. From its origins more than 30 years ago, surgical planning with analysis of flow hemodynamics and energy loss/efficiency has helped design and implement many modifications to existing techniques. On the basis of patient-specific medical imaging, surgical planning allows accurate model production that can then be manipulated in a virtual surgical environment, with the proposed solutions finally tested with advanced computational fluid dynamics to evaluate the results. Applications include a broad range of congenital heart disease, including patients with single-ventricle anatomy undergoing staged palliation, those with arch obstruction, with double outlet right ventricle, or with tetralogy of Fallot. In the present work, we focus on clinical applications of this exciting field. We describe the framework for these techniques, including brief descriptions of the engineering principles applied and the interaction between “benchtop” data with medical decision-making. We highlight some early insights learned from pioneers over the past few decades, including refinements in Fontan baffle geometries and configurations. Finally, we offer a glimpse into exciting advances that are presently being explored, including use of modelling for transcatheter interventions. In this era of personalized medicine, computational modelling and surgical planning allows patient-specific tailoring of interventions to optimize clinical outcomes.
La modélisation computationnelle et les simulations numériques sont de plus en plus utilisées pour planifier des interventions chirurgicales cardiovasculaires complexes chez les enfants et les jeunes adultes atteints de cardiopathie congénitale. Depuis ses origines, il y a plus de 30 ans, la planification chirurgicale par l'analyse de l'hémodynamique des écoulements et de l'efficacité ou des déperditions d'énergie des écoulements a contribué à concevoir et à mettre en œuvre de nombreuses modifications aux techniques existantes. En s'appuyant sur l'imagerie médicale se rapportant au patient, la planification chirurgicale permet la production de modèles précis qui peuvent ensuite être manipulés dans un environnement chirurgical virtuel. L’évaluation finale des résultats des solutions proposées est effectuée au moyen de tests de dynamique des fluides computationnelle avancée. Les applications touchent une large gamme de patients atteints de maladies cardiaques congénitales, dont les patients ayant un cœur univentriculaire et recevant des interventions palliatives en plusieurs temps, une obstruction de l'arc, un ventricule droit à double issue, ou encore une tétralogie de Fallot. Dans le présent travail, nous nous attardons aux applications cliniques de ce domaine passionnant. Nous décrivons le cadre dans lequel s'inscrivent ces techniques, et présentons notamment une brève description des principes d'ingénierie qui sont appliqués et de l'interaction entre les données de laboratoire et la prise de décision médicale. Nous mettons en relief quelques leçons tirées par les pionniers au cours des dernières décennies, dont les améliorations dans les géométries et les configurations des dérivations de type Fontan. Enfin, nous proposons un aperçu des passionnants progrès actuels, et notamment de l'utilisation de la modélisation pour les interventions transcathéter. À l'ère de la médecine personnalisée, la modélisation et la planification chirurgicale computationnelles rendent possibles des interventions personnalisées qui optimisent les résultats cliniques.
Technical feasibility and intermediate outcomes of using a handcrafted, area-preserving, bifurcated Y-graft modification of the Fontan procedure
2015, Journal of Thoracic and Cardiovascular Surgery
To demonstrate the technical feasibility and describe intermediate outcomes for the initial patients undergoing handcrafted, area-preserving, Y-graft modification of the Fontan procedure.
A retrospective review of a pilot study was undertaken to describe preoperative, intraoperative, and postoperative results.
Six patients underwent successful procedures and remain alive 3 to 4 years later. The median age at operation was 3.3 years, and median weight was 13.2 kg. Five operations were done without cardiopulmonary bypass and no intraoperative pressure gradients were found. Five patients were extubated by postoperative day 1, Fontan pressures were 12 to 14 mm Hg, transpulmonary gradients were 6 to 8 mm Hg, and no renal or hepatic function abnormalities were found. Length of stay was 10 to 64 days. One patient required venovenous extracorporeal membrane oxygenation for previously undiagnosed plastic bronchitis (64-day stay); another required reoperation for an incidentally diagnosed aortic thrombus (44-day stay). One patient had occlusion of a Y-graft limb noted on magnetic resonance imaging follow-up at 3 months. Catheterization showed excellent hemodynamic parameters and no Fontan obstruction. Occlusion was believed to be due to right-sided pulmonary arteriovenous malformations and widely discrepant flow (80%) to the right lung leading to low flow in the left limb.
The area-preserving, bifurcated Y-graft Fontan modification is technically feasible and shows excellent intermediate outcomes. Additional study is required to determine whether the advantages seen in the computational models will be realized in patients over the long-term, and to optimize patient selection for each of the various Fontan options now available.
Intraatrial conduit fontan procedure: Indications, operative techniques, and clinical outcomes
2015, Annals of Thoracic Surgery
Citation Excerpt :
Bleeding at the anastomosis between the prosthesis and autologous tissue is common, especially in cyanotic patients with thin and fragile pulmonary artery walls. Second, a flared junction design between the IAC and the pulmonary artery creates more efficient hemodynamics [15]. In recent years, computational fluid dynamics techniques have been applied to the design, refinement, and assessment of the Fontan procedure by calculating flow patterns and pressure changes within a virtual model of the Fontan pathway.
The intraatrial conduit (IAC) Fontan procedure is one of the Fontan modifications and is usually not considered the first choice. In this large series, we report our experience of the IAC Fontan procedure for the treatment of a functional single ventricle and review its indications, techniques, and clinical outcomes.
Between 2009 and 2013, 101 patients with a functional single ventricle underwent an IAC Fontan procedure. The median surgical age was 44 months and weight was 15 kg. The cardiac malformations included tricuspid atresia in 10, double-inlet left ventricle in 8, double-inlet right ventricle in 15, double-outlet right ventricle in 18, congenitally corrected transposition of the great arteries in 11, complete atrioventricular septal defect in 8; criss-cross in 1, pulmonary atresia with intact ventricular septum in 3, Ebstein anomaly in 3, and others in 3. Heterotaxy syndrome was found in 21 patients. An intraatrial polytetrafluoroethylene conduit was implanted to construct the Fontan pathway. In 75 patients, a 2.7-mm to 4.5-mm fenestration was made to reduce the intraconduit pressure. Eighty-three patients had previously undergone a Glenn operation. Median follow-up was 29 months (range, 2 to 60 months). Doppler echocardiography, electrocardiography, and Holter monitoring were used to evaluate hemodynamic performance and arrhythmias.
There were 2 hospital deaths, 1 Fontan takedown, and 3 midterm deaths. Conduit thrombosis developed in 1 patient. Atrial flutter developed in 2 patients, who underwent electrical cardioversion. Junctional bradycardia developed in 2 patients, but they did not require permanent pacemakers. Overall survival was 97.0% at 1 year and 94.1% at 5 years.
The IAC modification provides excellent operative and midterm outcomes in most patients with a functional single ventricle. However, a longer follow-up time is required to demonstrate its real advantages.
Alternative techniques for the fontan operation
2014, Operative Techniques in Thoracic and Cardiovascular Surgery

View all citing articles on Scopus

View full text

Original Articles: CardiovascularToward designing the optimal total cavopulmonary connection: an in vitro study

Abstract

Section snippets

Material and methods

Curved models

Curved models

Acknowledgements

J Thorac Cardiovasc Surg

J Am Coll Cardiol

J Thorac Cardiovasc Surg

Am J Cardiol

J Am Coll Cardiol

Surgical repair of tricuspid atresia

Thorax

Original Articles: Cardiovascular
Toward designing the optimal total cavopulmonary connection: an in vitro study