Ann Thorac Surg 1998;65:1741-1745
© 1998 The Society of Thoracic Surgeons
Original articles: cardiovascular
Superior Hydrodynamics of a Modified Cavopulmonary Connection for the Norwood Operation
Hans-Hinrich Sievers, MD, PhDa,
Anja Gerdes, MDa,
Jörg Kunze, MScb,
Gerd Pfister, PhDb
a Department of Cardiac Surgery, Medical University of Lübeck, Lübeck, Germany
b Institute of Applied Physics, Christian-Albrechts-University of Kiel, Kiel, Germany
Accepted for publication January 20, 1998.
Address reprint requests to Dr Sievers, Klinik für Herzchirurgie, Medizinische Universität zu Lübeck, Ratzeburger Allee 160, D-23538 Lübeck, Germany
e-mail: (herzchir{at}medinf.mu-luebeck.de)
 |
Abstract
|
|---|
Background. In the Fontan circulation, energy consumption at the cavopulmonary connection is crucial. Our hypothesis was that a modification of the standard Norwood variant of cavopulmonary connection with an extended anastomosis would improve hydrodynamics.
Methods. The in vitro hydrodynamics of two different Perspex glass models resembling the Norwood variant of cavopulmonary connection (model I) and the modification (model II) were analyzed in a mock circulation at nonpulsatile flows of 2 to 5 L/min to simulate rest and exercise. The pulmonary flow split was varied to imitate varying lung resistances. Inferior-to-superior caval flow ratio and size of models were increased to simulate growth.
Results. The pulmonary flow was preferentially directed to the left lung in model I and was better balanced in model II. Power losses increased exponentially with total flow in both models and were markedly higher in model I. These differences were attenuated in the larger models. Anastomotic turbulences were larger in model I. Power losses in both models were relatively insensitive to changes in pulmonary flow split.
Conclusions. The proposed modification of the Norwood variant of cavopulmonary connection seems to be hydrodynamically advantageous and warrants further evaluation.
|
Introduction
|
|---|
Hypoplastic left heart syndrome accounts for 9% of infants suffering from critical heart disease, with 95% dying in the first month of life [1]. In addition to cardiac allotransplantation [2], a physiologic repair of this heart defect became available with the introduction of the Norwood procedure [3, 4]. The concept of this conventional reconstructive surgical technique is in part based on a modification of the Fontan principle [5], aiming at bypassing the single pumping chamber. The energy source for propelling blood through the lungs in the Fontan circulation is limited [6]. Power losses at cavopulmonary connections interfere with forward flow and may impair clinical outcome [7]. In the Norwood variant of the Fontan procedure the systemic venous return is baffled to the right pulmonary artery using an augmentation gusset of pulmonary homograft material. The caval blood that flows into the right lung has to turn around a sharp nonstreamlined angulation of almost 180 degrees [4]. De Leval and colleagues [6], as well as Kim and associates [8], demonstrated unfavorable hydrodynamics on the downstream sides of sharply angled corners. Therefore the aim of this in vitro study was to evaluate by means of models the hydrodynamics of the standard Norwood variant of cavopulmonary connection and a modification of this technique reducing flow deflection (Figs 1, 2). By varying pulmonary flow split and total flow, as well as the ratio of inferior to superior caval flow and the size of the models, the effects of changing lung resistance and activity, as well as growth, were taken into consideration.
View larger version (29K):
[in this window]
[in a new window]
|
Fig 1. Model I simulating the standard Norwood variant of cavopulmonary connection (upper panel) and the proposed modification (lower panel, model II). (AG = augmentation gusset; IVC = inferior vena cava; RPA = right pulmonary artery; SVC = superior vena cava.)
|
|
View larger version (44K):
[in this window]
[in a new window]
|
Fig 2. Norwood variant of cavopulmonary connection (left) showing the venous blood flowing to the right pulmonary artery around the sharp angulation of the cavopulmonary junction and modification (right) with an enlarged anastomosis allowing the venous blood to flow to the right pulmonary artery around the smoothly angled cavopulmonary junction. (AG = augmentation gusset; RA = right atrium; RPA = right pulmonary artery; Shaded area = flow direction; SVC = superior vena cava.)
|
|
|
Material and methods
|
|---|
Configuration of models
The inlet duct of the models was drilled into a block of Perspex glass. The block was divided into two halves of equal size perpendicular to the inlet duct, and one half of the outlet duct was milled just below the inlet duct longitudinally into both corresponding surfaces of intersection. Additionally a triagonal piece of the Perspex glass with a semicircular roof that simulates the augmentation gusset in the standard Norwood procedure was excised out of both parts of the block between the left side of the inlet duct and the left outlet duct, to create the anastomosis. By combining the two mirror-imaged halves the model was completed. Two types of models were made. Model I (Fig 1) imitates the commonly performed Norwood variant of cavopulmonary connection (Fig 2). In model II (see Fig 1), which imitates the proposed modification (Fig 2), the anastomosis was extended far beneath the superior vena cava. The Norwood procedure is usually completed when the size of the pulmonary artery is approximately 10 mm in diameter, reflecting an age of approximately 2 to 3 years [9, 10]. Therefore, the diameters of the inlet and outlet ducts were set at 10 mm. To investigate the influence of growth a second series of models with the same configuration but with inlet and outlet ducts 12 mm in diameter were made, comparable with superior vena cava and right pulmonary artery diameters of children aged 4 to 6 years [9, 10]. The inlet ducts were designated as inferior and superior vena cava and the outlet ducts as right and left pulmonary artery.
Test circuit
The nonpulsatile flow test circuit consisted of two regulated direct-current pumps to supply the desired flow rates for the superior and inferior caval veins, the silicone connecting tubes of 12 mm in diameter, the models, and two overflow reservoirs connected to the pulmonary arteries. The pumps were positioned at a distance of 2 m from the models to assure laminar flow conditions. A mixture of 40% glycerin and 60% demineralized water was used as test fluid. To ensure a constant viscosity of 
= 3.6 · 10-3 Pa · s and a density of 
= 1,090 kg/m3 the temperature of the test fluid was maintained at 24°C by temperature control. The inferior-to-superior caval flow ratio was set at 1:1 for the 10-mm models and at 2:1 for the 12-mm models to adjust for growth, according to Salim and coworkers [11]. Pulmonary flow split was altered in the range of 20/80, 30/70, 40/60, 45/55, and 50/50 to simulate varying lung resistances. This was achieved by adjusting the head differential between the overflow reservoirs and the models.
Flows in the caval veins and the pulmonary arteries were measured simultaneously by four flow meters. Total flow was varied between 2 and 5 L/min to simulate rest and exercise. Pressures were measured continuously by standpipes at a distance of 20 diameters of the model ducts away from the models, to assure laminar flow conditions.
All measurements were repeated five times at steady-state conditions and the mean value taken. The accuracy of pressure measurement was calculated to be ±1%.
The complexity of quadruplets of data was reduced to scalar values by calculating the total power loss (Ploss). The kinetic power loss is insignificant for comparison of these models and can be discarded, as also reported by Sharma and associates [12]. Static power loss (Ploss, static) was calculated as follows from the different flows (Q) and pressures (p):
Flow visualization
For flow visualization small air bubbles were added to the glycerin-water solution and illuminated with a helium neon laser that was diffused with a cylindrical lens. Pictures of flow pathlines were taken with a camera using a black-and-white negative film and a shutter speed of 1/90th second. Experimental conditions and experimental set-up were the same as for the pressure measurements.
|
Results
|
|---|
Pulmonary flow distribution
While nullifying the head differential between both overflow reservoirs and thus simulating equal resistances in both lungs, a significant difference in the pulmonary flow ratio between models I and II was observed (Fig 3). With increasing total flow, the flow distribution between the right and left pulmonary arteries diverged increasingly. This effect was most dominant in model I.
View larger version (28K):
[in this window]
[in a new window]
|
Fig 3. In vitro pulmonary flow distribution in models I (A) and II (B). The diameters of inlet and outlet ducts in models I and II were 10 and 12 mm, respectively, and the flow ratio of inferior to superior vena cava was 1:1/2:1. (LPA = left pulmonary artery; mod. = model; RPA = right pulmonary artery.)
|
|
Power losses
With increasing total flow, power losses increased exponentially in both types and sizes of models (Fig 4). Generally power losses were lower in model II when compared with model I. Furthermore, the larger models provided fewer power losses than the smaller models.
View larger version (35K):
[in this window]
[in a new window]
|
Fig 4. In vitro power losses in models I (A) and II (B) in relation to changes in lung resistance simulated by pulmonary flow split, which is expressed as right pulmonary artery flow rate. The diameters of the inlet and outlet ducts in models I and II were 10 and 12 mm, respectively, and the flow ratio of inferior to superior vena cava was 1:1/2:1. (mod. = model; mW = milliwatt; RPA = right pulmonary artery.)
|
|
At 2, 3, 4, and 5 L/min the percentages of power losses were 17.60%, 27.49%, 36.03%, and 43.15%, respectively, for the 10-mm version of model I with balanced pulmonary flow, and 8.03%, 12.69%, 18.08%, and 22.92%, respectively, for the 10-mm version of model II. For the 12-mm version of model I with balanced pulmonary flow, the percentages of power losses at 2, 3, 4, and 5 L/min were 9.60%, 14.78%, 19.24%, and 22.61%, respectively, and 8.23%, 12.74%, 15.87%, and 18.52%, respectively, for the 12-mm version of model II. In both types of models power losses were lowest when the pulmonary flow was equally distributed and relatively insensitive to asymmetric pulmonary flow.
Flow visualization
The different flow patterns are depicted in Figures 5 and 6.
View larger version (54K):
[in this window]
[in a new window]
|
Fig 5. Photograph (A) and scheme (B) of the flow pattern at the cavopulmonary anastomosis in model I. In this cross-sectional view through the midline of the right pulmonary artery (RPA) the preferred, almost laminar, flow from the superior vena cava (SVC) to the left lung (arrows) and the nonstreamlined cavity beneath the sharply angled cavopulmonary junction is demonstrated.
|
|
View larger version (59K):
[in this window]
[in a new window]
|
Fig 6. Photograph (A) and scheme (B) of the flow pattern at the cavopulmonary anastomosis in model II. In comparison with model I (see Fig 5), a more spread and streamlined flow pattern with fewer turbulences is depicted. (RPA = right pulmonary artery; SVC = superior vena cava.)
|
|
|
Comment
|
|---|
In the Fontan circulation the power source for propelling blood through the pulmonary circulation is rather limited. The right atrial contraction does not provide additional power [6] and the systemic venous blood has to pass the pulmonary vascular bed driven by a residue of only 10% of left ventricular energy [6]. Thus, every effort to reduce energy dissipation and thereby resistance to forward flow, especially at the surgically susceptible cavopulmonary connections, has to be exercised to improve outcome [7].
In the standard Norwood variant of cavopulmonary connection the venous blood directed to the right pulmonary artery has to surmount a sharp angulation of almost 180 degrees (see Figs 2, 5). This flow deflection around a sharply contoured corner causes flow separation, flow disturbances, and localized turbulence, increasing maldistribution of pulmonary flow, as shown in model I.
Thus, the caval blood entering the anastomosis preferentially flows into the left pulmonary artery (see Fig 3), underlined by a streamlined flow pattern in this direction (see Fig 5). In model II flow deflection to the right pulmonary artery is reduced to almost 90 degrees (see Figs 2, 6), resulting in a more balanced flow distribution between both pulmonary arteries (see Fig 3). With the larger-sized models simulating growth, the difference in pulmonary flow distribution between models I and II is attenuated. It is well known that turbulence constitutes the main source of power dissipation [6]. This may explain the high power losses observed in our study, especially in the small model I at high total flow (see Fig 4A), which goes together with the largest area of turbulence within the anastomosis (Fig 5). In this setting simulating exercise, the power losses were more than twice as small in model II as in model I (see Fig 4A). This is accompanied by a reduction of localized turbulence and flow separation on the downstream side of the cavopulmonary junction (see Fig 6). The difference in the power losses between models I and II was attenuated with larger-sized models (see Fig 4B), underscoring the need for preserving the growth potential of the anastomosis by using interrupted sutures or resorbable suture material. The present study further demonstrates that pulmonary flow split simulating lung resistance did not essentially influence power losses (see Fig 4). This is of particular clinical significance inasmuch as during the postoperative period the bilateral balance of lung resistance may change, temporarily causing maldistribution of lung perfusion, however, most likely without a significant increase in power losses.
Our rather simple in vitro model is far from resembling the in vivo Fontan circulation. Only the hydrodynamics at the surgically susceptible anastomosis were evaluated, with rigid tubes not simulating vascular compliance. Other determinants for energy loss in the Fontan circulation, such as pulmonal arterial compliance, ventricular performance, left atrial pressure, and especially ventricular compliance, were not investigated. We could, however, provide some evidence that the proposed modification of the Norwood variant of cavopulmonary anastomoses probably provides the advantage of a better balanced pulmonary flow distribution and less power consumption, predominantly at a young age and during exercise, warranting further in vivo evaluation.
|
Acknowledgments
|
|---|
We are indebted to Ms Bettina Hansen and Ms Kirsti Scheffel for expert assistance in preparation of the manuscript.
|
References
|
|---|
- Fyler D.C. Report of the New England Regional Infant Cardiac Program. Pediatrics 1980;65(Suppl):377-461.[Abstract]
- Bailey L.L., Nehlsen-Cannarella S.L., Doroshow R.W., et al. Cardiac allotransplantation in newborns as therapy for hypoplastic left heart syndrome. N Engl J Med 1986;315:949-951.[Medline]
- Norwood W.I., Lang P., Hansen D.D. Physiologic repair of aortic atresia-hypoplastic left heart syndrome. N Engl J Med 1983;308:23-26.[Medline]
- Norwood W.I., Jacobs M.L., Murphy J.D. Fontan procedure for hypoplastic left heart syndrome. Ann Thorac Surg 1992;54:1025-1030.[Abstract]
- Fontan F., Baudet E. Surgical repair of tricuspid atresia. Thorax 1971;26:240-248.[Abstract/Free Full Text]
- De Leval M.R., Kilner P., Gewillig M., Bull C. Total cavopulmonary connection: a logical alternative to atriopulmonary connection for complex Fontan operations. Thorac Cardiovasc Surg 1988;96:682-695.
- Kirklin J.W., Fernandez G., Fontan F., Naftel D.C., Ebner A., Blackstone E.H. Therapeutic use of right atrial pressures early after the Fontan operation. Eur J Cardiothorac Surg 1990;4:2-7.[Abstract]
- Kim Y.H., Walker P.G., Fontaine A.A., et al. Hemodynamics of the Fontan connection: an in vitro study. J Biomech Eng 1995;117:423-428.[Medline]
- Sievers H.H., Onnasch D.G.W., Lange P.E., Bernhard A., Heintzen P. Dimensions of the great arteries, semilunar valve roots and right ventricular outflow tract during growth: normative angiocardiographic data. Ped Cardiol 1983;4:189-196.
- Köttgen U., Bolt W., Kreislauf In: Brock J., ed. Biologische Daten für den Kinderarzt. Berlin: Springer, 1954:365.
- Salim M.A., DiSessa T.G., Arheart K.L., Alpert B.S. Contribution of superior vena caval flow to total cardiac output in children. A Doppler echocardiographic study. Circulation 1995;92:1860-1865.[Abstract/Free Full Text]
- Sharma S., Goudy S., Walker P., 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;27:1264-1269.[Abstract]
This article has been cited by other articles:
|
|
|
|
 
E. L. Bove, M. R. de Leval, F. Migliavacca, G. Guadagni, and G. Dubini
Computational fluid dynamics in the evaluation of hemodynamic performance of cavopulmonary connections after the norwood procedure for hypoplastic left heart syndrome
J. Thorac. Cardiovasc. Surg.,
October 1, 2003;
126(4):
1040 - 1047.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
